技术领域Technical Field
本申请涉及定位及图像处理领域，尤其涉及一种点云地图更新的方法及设备。The present application relates to the field of positioning and image processing, and in particular to a method and device for updating a point cloud map.
背景技术Background technique
视觉定位系统(visual positioning system，VPS)是基于终端设备采集的图像，对终端设备进行定位的技术。例如，终端设备采集周围环境获得第一图像，通过将第一图像与数据库中的预存的图像匹配，得到数据库中与第一图像匹配度最高的第二图像，根据第二图像的拍摄位置确定当前终端设备所在位置。Visual positioning system (VPS) is a technology for locating terminal devices based on images collected by terminal devices. For example, the terminal device collects the surrounding environment to obtain a first image, and then matches the first image with a pre-stored image in a database to obtain a second image in the database with the highest matching degree with the first image, and determines the current location of the terminal device based on the shooting position of the second image.
可以看出，VPS定位技术是以环境图像作为匹配粒度进行定位的，并且随着图像拍摄的时间、频次、角度等的调整，图像可以实现及时准确地反映现实世界环境的变化。而该及时准确的环境变化数据正是用于地图更新的关键数据。因此，如何将VPS定位技术应用于地图更新，提高地图更新的准确性和时效性，成为值得关注的问题。It can be seen that VPS positioning technology uses environmental images as the matching granularity for positioning, and with the adjustment of the time, frequency, angle, etc. of image shooting, the image can timely and accurately reflect the changes in the real world environment. And this timely and accurate environmental change data is the key data for map updates. Therefore, how to apply VPS positioning technology to map updates and improve the accuracy and timeliness of map updates has become a problem worthy of attention.
发明内容Summary of the invention
本申请实施例提供了一种点云地图更新的方法及设备，通过借助时间维度和新旧环境图像之间的特征匹配关系及不同类型图像的数量信息，解决点云地图更新效率低的问题。The embodiments of the present application provide a method and device for updating a point cloud map, which solve the problem of low efficiency in updating a point cloud map by leveraging the time dimension and the feature matching relationship between new and old environmental images and the quantity information of different types of images.
第一方面，提供了一种点云地图更新的方法，应用于云端，所述方法包括：In a first aspect, a method for updating a point cloud map is provided, which is applied to a cloud, and the method comprises:
获取终端设备发送的图像信息，所述图像信息包括第一区域的图像和所述第一区域的图像对应的位置信息；Acquire image information sent by a terminal device, where the image information includes an image of a first area and position information corresponding to the image of the first area;
根据所述第一区域的图像对应的位置信息，通过视觉定位系统VPS获取所述图像的定位结果，其中，定位结果准确的图像为正常定位图像，定位结果不准确的图像为异常定位图像；According to the position information corresponding to the image of the first area, a positioning result of the image is obtained through a visual positioning system VPS, wherein an image with an accurate positioning result is a normal positioning image, and an image with an inaccurate positioning result is an abnormal positioning image;
根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图；Determining whether it is necessary to update the point cloud map corresponding to the first area according to the number of abnormal positioning images;
当需要更新所述第一区域对应的点云地图时，获取所述第一区域对应的场景图，所述场景图的节点为具有特征匹配关系的所述正常定位图像和所述异常定位图像，所述场景图的边为所述节点对应的图像之间匹配的特征数目；When it is necessary to update the point cloud map corresponding to the first area, a scene graph corresponding to the first area is obtained, wherein the nodes of the scene graph are the normal positioning image and the abnormal positioning image having a feature matching relationship, and the edges of the scene graph are the number of features matched between the images corresponding to the nodes;
根据所述场景图中所述正常定位图像的数量和所述异常定位图像的数量，更新所述第一区域对应的点云地图。According to the number of the normal positioning images and the number of the abnormal positioning images in the scene graph, the point cloud map corresponding to the first area is updated.
根据本实现方式提供的点云地图更新的方法，通过云端与终端设备的交互获取已建图区域的图像，并基于VPS定位技术判断对点云地图更新的必要性，能够实现无需过度耗费人力，不仅极大地节约了地图更新的成本，还更加智能化地提升了点云地图更新时机的准确性，保证了点云地图鲜度。此外，本方法在更新之前基于构建的场景图判断新旧部分的点云地图的融合性，能够保证更新部分更好地融合到原有的离线三维模型中，获取融合效果更好的点云地图。According to the point cloud map update method provided by this implementation, the image of the mapped area is obtained through the interaction between the cloud and the terminal device, and the necessity of updating the point cloud map is judged based on the VPS positioning technology, which can achieve the goal of not over-consuming manpower, greatly saving the cost of map updates, and more intelligently improving the accuracy of the timing of point cloud map updates, ensuring the freshness of point cloud maps. In addition, before updating, this method judges the fusion of the new and old parts of the point cloud map based on the constructed scene graph, which can ensure that the updated part is better integrated into the original offline three-dimensional model, and obtain a point cloud map with better fusion effect.
结合第一方面，在第一方面的某些实现方式中，所述方法还包括：In combination with the first aspect, in some implementations of the first aspect, the method further includes:
存储离线三维模型，所述离线三维模型对应的区域为已建图区域；其中，The offline three-dimensional model is stored, and the area corresponding to the offline three-dimensional model is the mapped area; wherein,
所述离线三维模型包括第一部分和其他部分，所述第一部分为未更新的所述第一区域对应的点云地图，所述其他部分为所述已建图区域中剩余其他区域对应的点云地图。The offline three-dimensional model includes a first part and other parts, the first part is a point cloud map corresponding to the first area that has not been updated, and the other parts are point cloud maps corresponding to other remaining areas in the mapped area.
在一种实现方式中，剩余其他区域是指已建图区域中除第一GPS区域之外的区域，例如可以对应下文图2中箭头所指的GPS区域之外的其它区域。In one implementation, the remaining other areas refer to areas in the mapped area other than the first GPS area, for example, areas other than the GPS area indicated by the arrow in FIG. 2 below.
在一种实现方式中，云端将已建图区域划分为多个小块的GPS区域，这些GPS区域包括第一GPS区域。其中，云端可以将这些小块的GPS区域作为点云地图更新的处理单元，避免一次性处理规模较大的点云地图更新数据，提高数据处理的灵活性和稳定性。In one implementation, the cloud divides the mapped area into multiple small GPS areas, including the first GPS area. The cloud can use these small GPS areas as processing units for point cloud map updates, avoiding processing large-scale point cloud map update data at one time, and improving the flexibility and stability of data processing.
结合第一方面，在第一方面的某些实现方式中，所述图像信息还包括所述其他区域的图像和所述其他区域的图像对应的位置信息，所述方法还包括：In combination with the first aspect, in some implementations of the first aspect, the image information further includes the image of the other area and position information corresponding to the image of the other area, and the method further includes:
根据所述图像信息中的位置信息和所述第一区域的位置，进行图像分类，获取属于所述第一区域的图像。Image classification is performed according to the position information in the image information and the position of the first area to obtain an image belonging to the first area.
结合第一方面，在第一方面的某些实现方式中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：In combination with the first aspect, in some implementations of the first aspect, obtaining whether it is necessary to update the point cloud map corresponding to the first area according to the number of abnormal positioning images specifically includes:
若所述异常定位图像的数量等于或大于第一阈值，则确定需要更新所述第一区域对应的点云地图。If the number of the abnormal positioning images is equal to or greater than a first threshold, it is determined that the point cloud map corresponding to the first area needs to be updated.
结合第一方面，在第一方面的某些实现方式中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：In combination with the first aspect, in some implementations of the first aspect, obtaining whether it is necessary to update the point cloud map corresponding to the first area according to the number of abnormal positioning images specifically includes:
若所述异常定位图像的数量等于或大于第一阈值，且在某个异常时间之后，所述异常定位图像的占比等于或大于第二阈值，则确定需要更新所述第一区域对应的点云地图。If the number of the abnormal positioning images is equal to or greater than a first threshold, and after a certain abnormal time, the proportion of the abnormal positioning images is equal to or greater than a second threshold, it is determined that the point cloud map corresponding to the first area needs to be updated.
结合第一方面，在第一方面的某些实现方式中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：In combination with the first aspect, in some implementations of the first aspect, obtaining whether it is necessary to update the point cloud map corresponding to the first area according to the number of abnormal positioning images specifically includes:
若所述异常定位图像的数量等于或大于第一阈值；以及，If the number of abnormal positioning images is equal to or greater than a first threshold; and,
在某个异常时间之后，所述异常定位图像的占比等于或大于第二阈值；以及，After a certain abnormal time, the proportion of the abnormal positioning images is equal to or greater than a second threshold; and
所述正常定位图像和所述异常定位图像在语义分割后的语义类别差距大于预设差距，则确定需要更新所述第一区域对应的点云地图。If the semantic category difference between the normal positioning image and the abnormal positioning image after semantic segmentation is greater than a preset difference, it is determined that the point cloud map corresponding to the first area needs to be updated.
结合第一方面，在第一方面的某些实现方式中，所述当需要更新所述第一区域对应的点云地图时，获取所述第一区域对应的场景图，具体包括：In combination with the first aspect, in some implementations of the first aspect, when it is necessary to update the point cloud map corresponding to the first area, obtaining the scene graph corresponding to the first area specifically includes:
对所述第一区域在异常时间之前的所述正常定位图像和所述异常时间之后的异常定位图像进行共视判断；Performing a common-view judgment on the normal positioning image of the first area before the abnormal time and the abnormal positioning image after the abnormal time;
将共视判断后与其他图像具有共同特征的图像作为所述节点，将两两图像之间匹配的特征数目作为所述两两图像对应节点之间的边，获取所述场景图。The images having common features with other images after common view determination are taken as the nodes, and the number of features matched between the two images is taken as the edges between the two corresponding nodes of the two images to obtain the scene graph.
结合第一方面，在第一方面的某些实现方式中，所述根据所述场景图中所述正常定位图像的数量和所述异常定位图像的数量，对所述第一区域对应的点云地图进行更新，具体包括：In combination with the first aspect, in some implementations of the first aspect, updating the point cloud map corresponding to the first area according to the number of the normal positioning images and the number of the abnormal positioning images in the scene graph specifically includes:
当所述场景图包括的所述正常定位图像的数量等于或大于第三阈值，以及所述场景图包括的异常定位图像的数量等于或大于第四阈值时，统计所述场景图包括的所述正常定位图像的占比是否等于或大于第五阈值；若所述场景图包括的所述正常定位图像的占比是否等于或大于第五阈值，则对所述第一区域对应的点云地图进行更新。When the number of normal positioning images included in the scene graph is equal to or greater than the third threshold, and the number of abnormal positioning images included in the scene graph is equal to or greater than the fourth threshold, it is counted whether the proportion of the normal positioning images included in the scene graph is equal to or greater than the fifth threshold; if the proportion of the normal positioning images included in the scene graph is equal to or greater than the fifth threshold, the point cloud map corresponding to the first area is updated.
结合第一方面，在第一方面的某些实现方式中，所述更新所述第一区域对应的点云地图，具体包括：In combination with the first aspect, in some implementations of the first aspect, updating the point cloud map corresponding to the first area specifically includes:
对所述第一区域对应的所述正常定位图像和所述异常定位图像均包括的同一位置作为特征点；The same position included in the normal positioning image and the abnormal positioning image corresponding to the first area is taken as a feature point;
对所述特征点进行三角化处理，获取与所述特征点对应的三维点云信息；Performing triangulation processing on the feature points to obtain three-dimensional point cloud information corresponding to the feature points;
在所述离线三维模型中，基于所述三维点云信息更新所述第一区域对应的点云地图。In the offline three-dimensional model, a point cloud map corresponding to the first area is updated based on the three-dimensional point cloud information.
结合第一方面，在第一方面的某些实现方式中，所述图像信息还包括时间信息；In combination with the first aspect, in some implementations of the first aspect, the image information further includes time information;
所述根据所述图像信息中的位置信息和所述第一区域的位置，进行图像分类，获取属于所述第一区域的图像，还包括：The performing image classification according to the position information in the image information and the position of the first area to obtain the image belonging to the first area also includes:
所述根据所述图像信息中时间信息，进行图像分类，获取属于所述第一区域的且在同一时间段内的图像。The image classification is performed according to the time information in the image information to obtain images belonging to the first area and within the same time period.
第二方面，提供了一种设备，包括：In a second aspect, a device is provided, comprising:
一个或多个处理器；one or more processors;
一个或多个存储器；one or more memories;
所述一个或多个存储器存储有一个或多个计算机程序，所述一个或多个计算机程序包括指令，当所述指令被所述一个或多个处理器执行时，使得所述设备执行如上述第一方面中任一项所述的方法。The one or more memories store one or more computer programs, and the one or more computer programs include instructions. When the instructions are executed by the one or more processors, the device performs the method as described in any one of the first aspects above.
第三方面，提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机可执行程序指令，所述计算机可执行程序指令在被计算机上运行时，使所述计算机执行如上述第一方面中任一项所述的方法。In a third aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable program instructions, and when the computer-executable program instructions are executed on a computer, the computer executes the method as described in any one of the above-mentioned first aspects.
第四方面，提供了一种计算机程序产品，所述计算机程序产品包括计算机程序代码，当所述计算机程序代码在计算机上运行时，使所述计算机执行如上述第一方面中任一实现方式所述的方法。In a fourth aspect, a computer program product is provided, the computer program product comprising a computer program code, and when the computer program code is executed on a computer, the computer is caused to execute the method as described in any implementation manner of the first aspect above.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
图1A和图1B所示，为本申请实施例分别提供的一种区域实景示意图和三维点云模型的示意图。FIG. 1A and FIG. 1B show a schematic diagram of a regional real scene and a schematic diagram of a three-dimensional point cloud model respectively provided in an embodiment of the present application.
图2为本申请实施例提供的一种对已建图区域划分GPS区域的示意图。FIG. 2 is a schematic diagram of dividing a mapped area into GPS areas according to an embodiment of the present application.
图3为本申请实施例提供的一种点云地图更新的方法适用的系统架构的示意图。FIG3 is a schematic diagram of a system architecture applicable to a method for updating a point cloud map provided in an embodiment of the present application.
图4为本申请实施例提供的一种终端设备100的结构示意图。FIG4 is a schematic diagram of the structure of a terminal device 100 provided in an embodiment of the present application.
图5为本申请实施例提供的一种终端设备100的软件结构框图。FIG5 is a software structure block diagram of a terminal device 100 provided in an embodiment of the present application.
图6为本申请实施例提供的一种云端200的结构示意图。FIG6 is a schematic diagram of the structure of a cloud 200 provided in an embodiment of the present application.
图7为本申请实施例提供的一种第一GPS区域对应的位置示意图。FIG. 7 is a schematic diagram of a location corresponding to a first GPS area provided in an embodiment of the present application.
图8A至图8D为本申请实施例提供的一些地图更新的方法在实现过程可能涉及到的场景示意图。8A to 8D are schematic diagrams of scenarios that may be involved in the implementation process of some map updating methods provided in embodiments of the present application.
图9为本申请实施例提供的一种点云地图更新的方法的示意性流程图。FIG9 is a schematic flowchart of a method for updating a point cloud map provided in an embodiment of the present application.
图10为本申请实施例提供的一种场景图对应的模型示意图。FIG10 is a schematic diagram of a model corresponding to a scene graph provided in an embodiment of the present application.
图11为本申请实施例提供的一种点云地图更新的方法的示意性流程图。FIG11 is a schematic flowchart of a method for updating a point cloud map provided in an embodiment of the present application.
具体实施方式Detailed ways
需要说明的是，本申请实施例的实施方式部分使用的术语仅用于对本申请的具体实施例进行解释，而非旨在限定本申请。在本申请实施例的描述中，除非另有说明，“/”表示或的意思，例如，A/B可以表示A或B；本文中的“和/或”仅仅是一种描述关联障碍物的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，在本申请实施例的描述中，除非另有说明，“多个”是指两个或多于两个，“至少一个”、“一个或多个”是指一个、两个或两个以上。It should be noted that the terms used in the implementation method part of the embodiments of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application. In the description of the embodiments of the present application, unless otherwise specified, "/" means or, for example, A/B can mean A or B; "and/or" in this article is only a description of the association relationship of associated obstacles, indicating that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two, "at least one" and "one or more" mean one, two or more than two.
以下，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”特征可以明示或者隐含地包括一个或者更多个该特征。In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, it is defined that the "first" and "second" features may explicitly or implicitly include one or more of the features.
在本说明书中描述的参考“一个实施例”或“一些实施例”等意味着在本申请的一个或多个实施例中包括结合该实施例描述的特定特征、结构或特点。由此，在本说明书中的不同之处出现的语句“在一个实施例中”、“在一些实施例中”、“在其它一些实施例中”、“在另外一些实施例中”等不是必然都参考相同的实施例，而是意味着“一个或多个但不是所有的实施例”，除非是以其它方式另外特别强调。术语“包括”、“包含”、“具有”及它们的变形都意味着“包括但不限于”，除非是以其它方式另外特别强调。References to "one embodiment" or "some embodiments" etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present application. Thus, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
众所周知，随着时间的发展，我们生活的环境会不断发生变化。小到某建筑物上增加广告牌、马路旁增设路标；大到新建某个地标性建筑、道路的重新规划等。我们生活的真实世界的环境几乎每时每刻都在发生变化。As we all know, the environment we live in will continue to change over time. From small things like adding billboards to buildings and adding road signs to the side of the road to big things like building new landmark buildings and replanning roads. The real world environment we live in is changing almost every moment.
在很多情况下，地图已经成为辅助人们出行的重要工具，地图的准确性影响着人们出行的效率。如果在真实环境已经发生了变化，但地图却没有更新，就会严重影响用户体验。目前，点云地图(或称三维(3dimensions，3D)点云地图)产品的更新方式一般是当真实环境发生变化后，由人工现场进行地图检查，也就是需要安排专业人员重新采集环境信息，之后再对点云地图进行更新。然而，这种点云地图的更新方式不仅消耗人力，还会严重影响更新效率。In many cases, maps have become an important tool to assist people in traveling, and the accuracy of maps affects the efficiency of people's travel. If the real environment has changed but the map has not been updated, it will seriously affect the user experience. At present, the update method of point cloud map (or three-dimensional (3D) point cloud map) products is generally to conduct manual on-site map inspection when the real environment changes, that is, it is necessary to arrange professionals to re-collect environmental information and then update the point cloud map. However, this method of updating point cloud maps not only consumes manpower, but also seriously affects the update efficiency.
有鉴于此，本申请实施例提供了一种点云地图更新的方法及设备。该方法通过对终端设备上报的已建图区域的图像进行VPS定位，通过query图像与已建立的离线三维模型(也即点云地图对应模型)对应位置的图像进行特征匹配，获取其中定位准确的图像(下称正常定位图像)和定位不准确的图像(下称异常定位图像)。之后根据预设的第一条件判断是否需要对点云地图进行更新。若需要进行点云地图更新，则在更新过程中利用共视判断构建场景图，然后根据预设的第二条件判断更新部分可否较好地融入原有的离线三维模型。In view of this, an embodiment of the present application provides a method and device for updating a point cloud map. The method performs VPS positioning on the image of the mapped area reported by the terminal device, and performs feature matching between the query image and the image of the corresponding position of the established offline three-dimensional model (that is, the corresponding model of the point cloud map), thereby obtaining an image with accurate positioning (hereinafter referred to as a normal positioning image) and an image with inaccurate positioning (hereinafter referred to as an abnormal positioning image). Then, it is determined whether the point cloud map needs to be updated according to the preset first condition. If the point cloud map needs to be updated, a scene graph is constructed using common view judgment during the update process, and then it is determined whether the updated part can be better integrated into the original offline three-dimensional model according to the preset second condition.
与现有的人工现场检查地图不同，在本申请实施例提供的点云地图更新的方法中，一方面的改进包括：云端通过第一条件自主判断点云地图是否需要更新；另一方面的改进包括：通过对环境变化前后图像进行共视判断，构建场景图，之后通过第二条件判断更新部分与原有点云地图的融入效果。其中，第一条件可能与异常定位图像的数量，异常定位图像的占比随时间的变化，以及异常定位图像和正常定位图像的语义类别差距关联；第二条件可能与场景图中正常定位图像的数量，异常定位图像的数量以及正常定位图像的占比关联。关于第一条件和第二条件的具体内容将分别在下文实施例中进行介绍，此处暂不详述。Different from the existing manual on-site inspection maps, in the point cloud map update method provided in the embodiment of the present application, the improvements on the one hand include: the cloud autonomously determines whether the point cloud map needs to be updated through the first condition; the improvements on the other hand include: by making a common view judgment on the images before and after the environmental changes, a scene graph is constructed, and then the integration effect of the updated part with the original point cloud map is judged through the second condition. Among them, the first condition may be associated with the number of abnormal positioning images, the change of the proportion of abnormal positioning images over time, and the semantic category gap between abnormal positioning images and normal positioning images; the second condition may be associated with the number of normal positioning images in the scene graph, the number of abnormal positioning images and the proportion of normal positioning images. The specific contents of the first condition and the second condition will be introduced in the following embodiments respectively, and will not be described in detail here.
根据本申请实施例提供的点云地图更新的方法，通过云端与终端设备的交互获取已建图区域的图像，并基于VPS定位技术获取的异常定位图像的数量信息，判断对点云地图更新的必要性，能够实现无需过度耗费人力，不仅极大地节约了地图更新的成本，还更加智能化地提升了更新地图的时机的准确性，保持了地图鲜度。此外，本方法在最终更新点云地图之前，基于构建的场景图对点云地图更新融合性进行判断，能够保证更新部分更好地融合到原有的离线三维模型中，获取融合效果更好的点云地图。According to the method for updating the point cloud map provided in the embodiment of the present application, the image of the mapped area is obtained through the interaction between the cloud and the terminal device, and the necessity of updating the point cloud map is judged based on the number information of the abnormal positioning images obtained by the VPS positioning technology, which can achieve the goal of not over-consuming manpower, greatly saving the cost of map updates, and more intelligently improving the accuracy of the timing of map updates, thereby maintaining the freshness of the map. In addition, before the final update of the point cloud map, the method judges the fusion of the point cloud map update based on the constructed scene graph, which can ensure that the updated part is better integrated into the original offline three-dimensional model, and obtain a point cloud map with better fusion effect.
需要说明的是，本申请实施例提供的点云地图更新的方法可以适用于多种场景。示例性的，例如对任意点云地图进行更新的场景，或者基于3D模型构建的高精地图进行更新的场景等。也可以应用于虚拟现实(virtual reality，VR)/增强现实(augmentedreality，AR)场景下的点云地图更新，或者对智能交通(如无人驾驶)、智能导航中涉及的点云地图进行更新。本申请实施例对本申请提供的点云地图更新的方法应用的场景不作任何限定。It should be noted that the method for updating the point cloud map provided in the embodiment of the present application can be applicable to a variety of scenarios. For example, the scenario of updating any point cloud map, or the scenario of updating a high-precision map built based on a 3D model, etc. It can also be applied to updating point cloud maps in virtual reality (VR)/augmented reality (AR) scenarios, or to updating point cloud maps involved in intelligent transportation (such as unmanned driving) and intelligent navigation. The embodiment of the present application does not impose any limitation on the scenarios to which the method for updating the point cloud map provided in the present application can be applied.
为了更好地理解本申请实施例提供的点云地图更新的方法，以下首先对本文可能出现的一些技术术语或者定义进行介绍。In order to better understand the method for updating the point cloud map provided in the embodiment of the present application, some technical terms or definitions that may appear in this document are first introduced below.
1.已建图区域1. Mapped area
在本申请实施例中，对于某些区域，可以预先根据这些区域的形貌、布局等进行建模，构建该区域对应的三维模型，并将该三维模型存储至云端，本申请将已建有三维模型的区域定义为已建图区域。其中，这里的三维模型具体可以是离线三维模型，可对应于本申请中待更新的点云地图。In the embodiments of the present application, for certain areas, modeling can be performed in advance according to the morphology, layout, etc. of these areas, a 3D model corresponding to the area can be constructed, and the 3D model can be stored in the cloud. The present application defines the area with a 3D model as a mapped area. The 3D model here can specifically be an offline 3D model, which can correspond to the point cloud map to be updated in the present application.
2.视觉定位系统VPS2. Visual Positioning System VPS
VPS是指通过机器视觉来完成定位任务的系统。例如，在进行实景导航的过程中，将实时采集的图像与云端存储的地图进行匹配，以得到精确的定位。在本申请实施例中，用户可以通过具有VPS功能的终端设备采集图像，并将图像上传至云端。针对云端获取的这些图像(下称待匹配(query)图像)，云端可以基于query图像的位置信息在离线三维模型中检索与其对应的图像信息，之后对query图像和被检索到的图像进行特征匹配，获取该query图像的定位结果。VPS refers to a system that uses machine vision to complete positioning tasks. For example, in the process of real-life navigation, the real-time captured image is matched with the map stored in the cloud to obtain accurate positioning. In an embodiment of the present application, a user can capture images through a terminal device with a VPS function and upload the images to the cloud. For these images obtained by the cloud (hereinafter referred to as query images), the cloud can retrieve the corresponding image information in the offline three-dimensional model based on the location information of the query image, and then perform feature matching on the query image and the retrieved image to obtain the positioning result of the query image.
3.异常定位图像、正常定位图像3. Abnormal positioning image, normal positioning image
在本申请实施例中，云端基于VPS技术对query图像进行特征匹配之后，可以获取两种类型的图像，一种是定位结果准确的图像，本申请将其定义为正常定位图像；另一种是定位结果不准确的图像，本申请将其定义为异常定位图像。In an embodiment of the present application, after the cloud performs feature matching on the query image based on VPS technology, two types of images can be obtained. One is an image with accurate positioning results, which is defined as a normal positioning image in this application; the other is an image with inaccurate positioning results, which is defined as an abnormal positioning image in this application.
其中，定位结果准确可以指，query图像与离线三维模型中对应图像的图像匹配度等于或大于某个预设的第一数值(如75％、80％等)；或者，query图像与离线三维模型中对应图像的特征匹配数目较多，例如特征匹配数目等于或大于预设的第二数值。结果不准确可以指，query图像与离线三维模型中对应图像的图像匹配度小于预设的第一数值(如75％、80％等)；或者，query图像与离线三维模型中对应图像的特征匹配数目较少，例如特征匹配数目小于预设的第二数值。Among them, accurate positioning results may refer to that the image matching degree between the query image and the corresponding image in the offline 3D model is equal to or greater than a preset first value (such as 75%, 80%, etc.); or, the number of feature matches between the query image and the corresponding image in the offline 3D model is large, for example, the number of feature matches is equal to or greater than a preset second value. Inaccurate results may refer to that the image matching degree between the query image and the corresponding image in the offline 3D model is less than a preset first value (such as 75%, 80%, etc.); or, the number of feature matches between the query image and the corresponding image in the offline 3D model is small, for example, the number of feature matches is less than a preset second value.
可以理解的，正常定位图像可以表示其对应的真实环境与对该区域建模时的一致。假如云端获取的query图像全部与离线三维模型中分别对应的图像实现正常的图像匹配，也即query图像均为正常定位图像，这说明真实环境没有发生变化，此时也就无需对该区域的点云地图进行更新。异常定位图像可以表示其对应的真实环境与对该区域建模时的不一致，说明该区域的真实环境发生变化，此时可能需要对该区域的点云地图进行更新。It can be understood that a normal positioning image can indicate that the corresponding real environment is consistent with the modeling of the area. If all query images obtained from the cloud achieve normal image matching with the corresponding images in the offline 3D model, that is, the query images are all normal positioning images, this means that the real environment has not changed, and there is no need to update the point cloud map of the area. An abnormal positioning image can indicate that the corresponding real environment is inconsistent with the modeling of the area, indicating that the real environment of the area has changed, and the point cloud map of the area may need to be updated.
4.全球定位系统(global positioning system，GPS)区域/网格区域4. Global Positioning System (GPS) Area/Grid Area
GPS区域(或称网格区域)是指为了避免一次性处理规模较大的点云地图更新数据，而将面积较大的已建图区域划分为多个小块区域。云端可以将这些小块的GPS区域作为处理单元，减少一次性处理的数据量。通常情况下，可以根据GPS信息对已建图区域进行划分，故在本申请中将这些小块区域定义为GPS区域(或网格区域)。GPS area (or grid area) refers to the division of a large mapped area into multiple small areas in order to avoid processing large-scale point cloud map update data at one time. The cloud can use these small GPS areas as processing units to reduce the amount of data processed at one time. Usually, the mapped area can be divided according to GPS information, so these small areas are defined as GPS areas (or grid areas) in this application.
以下结合附图，以某一已建图区域为例，对本申请实施例提供的点云地图更新的方法的实现过程进行场景化介绍。In the following, in combination with the accompanying drawings, taking a certain mapped area as an example, the implementation process of the method for updating the point cloud map provided in the embodiment of the present application is introduced in a scenario-based manner.
示例性的，如图1A和图1B所示，分别为本申请实施例提供的一种区域实景示意图和对应的三维模型的示意图。Exemplarily, as shown in FIG. 1A and FIG. 1B , they are respectively a schematic diagram of a regional real scene and a schematic diagram of a corresponding three-dimensional model provided in an embodiment of the present application.
在一些实施例中，云端可以预先获取已建图区域对应的三维模型。云端获取该三维模型的方式不限，比如云端可以自己对已建图区域建模，或者云端也可以从第三方获取该三维模型。其中，这里的三维模型可以采用现有的方式获取，比如通过激光雷达采集环境中的三维点云数据，基于这些三维点云数据生成三维模型，本申请实施例对此不作限定。In some embodiments, the cloud can pre-acquire a three-dimensional model corresponding to the mapped area. The cloud can obtain the three-dimensional model in any manner, such as the cloud can model the mapped area itself, or the cloud can obtain the three-dimensional model from a third party. The three-dimensional model here can be obtained in an existing manner, such as by collecting three-dimensional point cloud data in the environment through a laser radar, and generating a three-dimensional model based on these three-dimensional point cloud data, which is not limited in the embodiments of the present application.
在一些实施例中，云端可以将三维模型存储至数据库，获取离线三维模型。In some embodiments, the cloud can store the three-dimensional model in a database to obtain an offline three-dimensional model.
示例性的，如图2所示，为本申请实施例提供的一种对已建图区域划分GPS区域的示意图。Exemplarily, as shown in FIG2 , it is a schematic diagram of dividing a mapped area into GPS areas provided in an embodiment of the present application.
在一些实施例中，云端可以基于已建图区域的GPS信息，将其划分为多个网格区域。例如，在已建图区域的GPS地图中，根据GPS位置将其划分为多个GPS区域。云端可以存储各个GPS区域的位置信息，如GPS区域对应的GPS范围值、顶点位置、尺寸大小等等。In some embodiments, the cloud can divide the mapped area into multiple grid areas based on GPS information. For example, in a GPS map of the mapped area, it is divided into multiple GPS areas based on GPS location. The cloud can store location information of each GPS area, such as GPS range value, vertex location, size, etc. corresponding to the GPS area.
示例性的，如图3所示，为本申请实施例提供的一种点云地图更新的方法适用的系统架构的示意图。该系统架构包括至少一个终端设备100和云端200。As shown in Fig. 3, it is a schematic diagram of a system architecture applicable to a method for updating a point cloud map provided in an embodiment of the present application. The system architecture includes at least one terminal device 100 and a cloud 200.
作为示例，图3以终端设备100是手机为例进行介绍。但在实际应用中，终端设备100可以是具有VPS功能及通信功能的多种类型的电子设备。例如，平板电脑、可穿戴设备、数码相机、上网本、个人数字助理(personal digital assistant，PDA)等。本申请实施例对终端设备100的类型不做任何限制。As an example, FIG3 introduces the terminal device 100 as a mobile phone. However, in actual applications, the terminal device 100 can be various types of electronic devices with VPS functions and communication functions. For example, a tablet computer, a wearable device, a digital camera, a netbook, a personal digital assistant (PDA), etc. The embodiment of the present application does not impose any restrictions on the type of the terminal device 100.
在一些实施例中，终端设备100可以用于采集已建图区域中的图像。此外，终端设备100还可以获取位置信息，如通过GPS获取位置坐标。可选地，终端设备100还可以获取所采集的图像对应的时间信息，如终端设备100可以在图像信息中添加时间戳。示例性的，图像对应的时间信息例如可以是终端设备100采集图像时的系统时间，或者是终端设备100自带的计时器时间，再或者是终端设备100从第三方设备获取的时间等等，本申请实施例对此不作限定。In some embodiments, the terminal device 100 can be used to collect images in a mapped area. In addition, the terminal device 100 can also obtain location information, such as obtaining location coordinates through GPS. Optionally, the terminal device 100 can also obtain time information corresponding to the collected image, such as the terminal device 100 can add a timestamp to the image information. Exemplarily, the time information corresponding to the image can be, for example, the system time when the terminal device 100 collects the image, or the timer time of the terminal device 100, or the time obtained by the terminal device 100 from a third-party device, etc., which is not limited in the embodiments of the present application.
在一些实施例中，终端设备100可以通过无线通信链路与云端200实现信息交互，如将图像信息上传至云端。其中，该图像信息可以包括终端设备100通过VPS功能采集的图像(即query图像)、图像对应的时间信息以及图像对应的位置信息等。终端设备100与云端200之间的无线通信链路对应的网络类型可以是基于现有通信技术标准的无线网络，如长期演进(long term evolution，LTE)无线网络、车联网(vehical to everything，V2X)、第五代移动通信技术(the 5th generation，5G)，以及未来有可能出现的其他类型无线网络，如第六代移动通信技术(the 6th generation，6G)等等。In some embodiments, the terminal device 100 can realize information interaction with the cloud 200 through a wireless communication link, such as uploading image information to the cloud. The image information may include an image (i.e., a query image) collected by the terminal device 100 through the VPS function, time information corresponding to the image, and location information corresponding to the image. The network type corresponding to the wireless communication link between the terminal device 100 and the cloud 200 can be a wireless network based on existing communication technology standards, such as a long term evolution (LTE) wireless network, a vehicle to everything (V2X), the fifth generation of mobile communication technology (5G), and other types of wireless networks that may appear in the future, such as the sixth generation of mobile communication technology (6G), etc.
在一些实施例中，云端200可以是一个服务器或者包括多个服务器的服务器集群。云端200具有多种能力，如无线通信能力、存储能力以及基于VPS的计算能力等等。举例来说，云端200可以用于存储终端设备上传的query图像，并对图像和离线三维模型进行特征匹配，然后基于匹配结果执行判断点云地图是否需要更新的操作，以及执行对点云地图的更新流程等。此外，云端200还可以对接收到的query图像进行分类。例如，根据位置信息对图像进行分类，将同一GPS区域中的图像被划分为同一类；再例如，根据时间信息对图像进行分类，如将同一GPS区域中的图像按照季节分类。In some embodiments, the cloud 200 can be a server or a server cluster including multiple servers. The cloud 200 has multiple capabilities, such as wireless communication capabilities, storage capabilities, and VPS-based computing capabilities, etc. For example, the cloud 200 can be used to store query images uploaded by terminal devices, and perform feature matching between images and offline three-dimensional models, and then perform operations to determine whether the point cloud map needs to be updated based on the matching results, and execute the update process of the point cloud map, etc. In addition, the cloud 200 can also classify the received query images. For example, images are classified according to location information, and images in the same GPS area are divided into the same category; for another example, images are classified according to time information, such as classifying images in the same GPS area according to seasons.
示例性的，如图4所示，为本申请实施例提供的一种终端设备100的示意性结构图。Exemplarily, as shown in FIG4 , there is a schematic structural diagram of a terminal device 100 provided in an embodiment of the present application.
终端设备100可以包括处理器110，外部存储器接口120，内部存储器121，通用串行总线(universal serial bus，USB)接口130，充电管理模块140，电源管理模块141，电池142，天线1，天线2，移动通信模块150，无线通信模块160，音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，传感器模块180，按键190，马达191，指示器192，摄像头193，显示屏194，以及用户标识模块(subscriber identification module，SIM)卡接口195等。其中传感器模块180可以包括压力传感器180A，陀螺仪传感器180B，气压传感器180C，磁传感器180D，加速度传感器180E，距离传感器180F，接近光传感器180G，指纹传感器180H，温度传感器180J，触摸传感器180K，环境光传感器180L，骨传导传感器180M等。The terminal device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.
可以理解的是，本发明实施例示意的结构并不构成对终端设备100的具体限定。在本申请另一些实施例中，终端设备100可以包括比图示更多或更少的部件，或者组合某些部件，或者拆分某些部件，或者不同的部件布置。图示的部件可以以硬件，软件或软件和硬件的组合实现。It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the terminal device 100. In other embodiments of the present application, the terminal device 100 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.
处理器110可以包括一个或多个处理单元，例如：处理器110可以包括应用处理器(application processor，AP)，调制解调处理器，图形处理器(graphics processingunit，GPU)，图像信号处理器(image signal processor，ISP)，控制器，存储器，视频编解码器，数字信号处理器(digital signal processor，DSP)，基带处理器，和/或神经网络处理器(neural-network processing unit，NPU)等。其中，不同的处理单元可以是独立的器件，也可以集成在一个或多个处理器中。The processor 110 may include one or more processing units, for example, the processor 110 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and/or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.
其中，控制器可以是终端设备100的神经中枢和指挥中心。控制器可以根据指令操作码和时序信号，产生操作控制信号，完成取指令和执行指令的控制。The controller may be the nerve center and command center of the terminal device 100. The controller may generate an operation control signal according to the instruction operation code and the timing signal to complete the control of fetching and executing instructions.
处理器110中还可以设置存储器，用于存储指令和数据。在一些实施例中，处理器110中的存储器为高速缓冲存储器。该存储器可以保存处理器110刚用过或循环使用的指令或数据。如果处理器110需要再次使用该指令或数据，可从所述存储器中直接调用。避免了重复存取，减少了处理器110的等待时间，因而提高了系统的效率。The processor 110 may also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. The memory may store instructions or data that the processor 110 has just used or cyclically used. If the processor 110 needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.
在一些实施例中，处理器110可以包括一个或多个接口。接口可以包括集成电路(inter-integrated circuit，I2C)接口，集成电路内置音频(inter-integrated circuitsound，I2S)接口，脉冲编码调制(pulse code modulation，PCM)接口，通用异步收发传输器(universal asynchronous receiver/transmitter，UART)接口，移动产业处理器接口(mobile industry processor interface，MIPI)，通用输入输出(general-purposeinput/output，GPIO)接口，用户标识模块(subscriber identity module，SIM)接口，和/或通用串行总线(universal serial bus，USB)接口等。In some embodiments, the processor 110 may include one or more interfaces. The interface may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver/transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input/output (GPIO) interface, a subscriber identity module (SIM) interface, and/or a universal serial bus (USB) interface, etc.
I2C接口是一种双向同步串行总线，包括一根串行数据线(serial data line，SDA)和一根串行时钟线(derail clock line，SCL)。在一些实施例中，处理器110可以包含多组I2C总线。处理器110可以通过不同的I2C总线接口分别耦合触摸传感器180K，充电器，闪光灯，摄像头193等。例如：处理器110可以通过I2C接口耦合触摸传感器180K，使处理器110与触摸传感器180K通过I2C总线接口通信，实现终端设备100的触摸功能。The I2C interface is a bidirectional synchronous serial bus, including a serial data line (SDA) and a serial clock line (SCL). In some embodiments, the processor 110 may include multiple groups of I2C buses. The processor 110 may be coupled to the touch sensor 180K, the charger, the flash, the camera 193, etc. through different I2C bus interfaces. For example: the processor 110 may be coupled to the touch sensor 180K through the I2C interface, so that the processor 110 communicates with the touch sensor 180K through the I2C bus interface, thereby realizing the touch function of the terminal device 100.
I2S接口可以用于音频通信。在一些实施例中，处理器110可以包含多组I2S总线。处理器110可以通过I2S总线与音频模块170耦合，实现处理器110与音频模块170之间的通信。在一些实施例中，音频模块170可以通过I2S接口向无线通信模块160传递音频信号，实现通过蓝牙耳机接听电话的功能。The I2S interface can be used for audio communication. In some embodiments, the processor 110 can include multiple I2S buses. The processor 110 can be coupled to the audio module 170 via the I2S bus to achieve communication between the processor 110 and the audio module 170. In some embodiments, the audio module 170 can transmit an audio signal to the wireless communication module 160 via the I2S interface to achieve the function of answering a call through a Bluetooth headset.
PCM接口也可以用于音频通信，将模拟信号抽样，量化和编码。在一些实施例中，音频模块170与无线通信模块160可以通过PCM总线接口耦合。在一些实施例中，音频模块170也可以通过PCM接口向无线通信模块160传递音频信号，实现通过蓝牙耳机接听电话的功能。所述I2S接口和所述PCM接口都可以用于音频通信。The PCM interface can also be used for audio communication, sampling, quantizing and encoding analog signals. In some embodiments, the audio module 170 and the wireless communication module 160 can be coupled via a PCM bus interface. In some embodiments, the audio module 170 can also transmit audio signals to the wireless communication module 160 via the PCM interface to realize the function of answering calls via a Bluetooth headset. Both the I2S interface and the PCM interface can be used for audio communication.
UART接口是一种通用串行数据总线，用于异步通信。该总线可以为双向通信总线。它将要传输的数据在串行通信与并行通信之间转换。在一些实施例中，UART接口通常被用于连接处理器110与无线通信模块160。例如：处理器110通过UART接口与无线通信模块160中的蓝牙模块通信，实现蓝牙功能。在一些实施例中，音频模块170可以通过UART接口向无线通信模块160传递音频信号，实现通过蓝牙耳机播放音乐的功能。The UART interface is a universal serial data bus for asynchronous communication. The bus can be a bidirectional communication bus. It converts the data to be transmitted between serial communication and parallel communication. In some embodiments, the UART interface is generally used to connect the processor 110 and the wireless communication module 160. For example, the processor 110 communicates with the Bluetooth module in the wireless communication module 160 through the UART interface to implement the Bluetooth function. In some embodiments, the audio module 170 can transmit an audio signal to the wireless communication module 160 through the UART interface to implement the function of playing music through a Bluetooth headset.
MIPI接口可以被用于连接处理器110与显示屏194，摄像头193等外围器件。MIPI接口包括摄像头串行接口(camera serial interface，CSI)，显示屏串行接口(displayserial interface，DSI)等。在一些实施例中，处理器110和摄像头193通过CSI接口通信，实现终端设备100的拍摄功能。处理器110和显示屏194通过DSI接口通信，实现终端设备100的显示功能。The MIPI interface can be used to connect the processor 110 with peripheral devices such as the display screen 194 and the camera 193. The MIPI interface includes a camera serial interface (CSI), a display serial interface (DSI), etc. In some embodiments, the processor 110 and the camera 193 communicate via the CSI interface to implement the shooting function of the terminal device 100. The processor 110 and the display screen 194 communicate via the DSI interface to implement the display function of the terminal device 100.
GPIO接口可以通过软件配置。GPIO接口可以被配置为控制信号，也可被配置为数据信号。在一些实施例中，GPIO接口可以用于连接处理器110与摄像头193，显示屏194，无线通信模块160，音频模块170，传感器模块180等。GPIO接口还可以被配置为I2C接口，I2S接口，UART接口，MIPI接口等。The GPIO interface can be configured by software. The GPIO interface can be configured as a control signal or as a data signal. In some embodiments, the GPIO interface can be used to connect the processor 110 with the camera 193, the display 194, the wireless communication module 160, the audio module 170, the sensor module 180, etc. The GPIO interface can also be configured as an I2C interface, an I2S interface, a UART interface, a MIPI interface, etc.
USB接口130是符合USB标准规范的接口，具体可以是Mini USB接口，Micro USB接口，USB Type C接口等。USB接口130可以用于连接充电器为终端设备100充电，也可以用于终端设备100与外围设备之间传输数据。也可以用于连接耳机，通过耳机播放音频。该接口还可以用于连接其他终端，例如AR设备等。The USB interface 130 is an interface that complies with the USB standard specification, and specifically can be a Mini USB interface, a Micro USB interface, a USB Type C interface, etc. The USB interface 130 can be used to connect a charger to charge the terminal device 100, and can also be used to transmit data between the terminal device 100 and peripheral devices. It can also be used to connect headphones to play audio through the headphones. The interface can also be used to connect other terminals, such as AR devices, etc.
可以理解的是，本发明实施例示意的各模块间的接口连接关系，只是示意性说明，并不构成对终端设备100的结构限定。在本申请另一些实施例中，终端设备100也可以采用上述实施例中不同的接口连接方式，或多种接口连接方式的组合。It is understandable that the interface connection relationship between the modules illustrated in the embodiment of the present invention is only a schematic illustration and does not constitute a structural limitation on the terminal device 100. In other embodiments of the present application, the terminal device 100 may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.
充电管理模块140用于从充电器接收充电输入。其中，充电器可以是无线充电器，也可以是有线充电器。在一些有线充电的实施例中，充电管理模块140可以通过USB接口130接收有线充电器的充电输入。在一些无线充电的实施例中，充电管理模块140可以通过终端设备100的无线充电线圈接收无线充电输入。充电管理模块140为电池142充电的同时，还可以通过电源管理模块141为终端供电。The charging management module 140 is used to receive charging input from a charger. The charger may be a wireless charger or a wired charger. In some wired charging embodiments, the charging management module 140 may receive charging input from a wired charger through the USB interface 130. In some wireless charging embodiments, the charging management module 140 may receive wireless charging input through a wireless charging coil of the terminal device 100. While the charging management module 140 is charging the battery 142, it may also power the terminal through the power management module 141.
电源管理模块141用于连接电池142，充电管理模块140与处理器110。电源管理模块141接收电池142和/或充电管理模块140的输入，为处理器110，内部存储器121，外部存储器，显示屏194，摄像头193，和无线通信模块160等供电。电源管理模块141还可以用于监测电池容量，电池循环次数，电池健康状态(漏电，阻抗)等参数。在其他一些实施例中，电源管理模块141也可以设置于处理器110中。在另一些实施例中，电源管理模块141和充电管理模块140也可以设置于同一个器件中。The power management module 141 is used to connect the battery 142, the charging management module 140 and the processor 110. The power management module 141 receives input from the battery 142 and/or the charging management module 140, and supplies power to the processor 110, the internal memory 121, the external memory, the display screen 194, the camera 193, and the wireless communication module 160. The power management module 141 can also be used to monitor parameters such as battery capacity, battery cycle number, battery health status (leakage, impedance), etc. In some other embodiments, the power management module 141 can also be set in the processor 110. In other embodiments, the power management module 141 and the charging management module 140 can also be set in the same device.
终端设备100的无线通信功能可以通过天线1，天线2，移动通信模块150，无线通信模块160，调制解调处理器以及基带处理器等实现。The wireless communication function of the terminal device 100 can be implemented through the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor and the baseband processor.
天线1和天线2用于发射和接收电磁波信号。终端设备100中的每个天线可用于覆盖单个或多个通信频带。不同的天线还可以复用，以提高天线的利用率。例如：可以将天线1复用为无线局域网的分集天线。在另外一些实施例中，天线可以和调谐开关结合使用。Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in terminal device 100 can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization of antennas. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.
移动通信模块150可以提供应用在终端设备100上的包括2G/3G/4G/5G等无线通信的解决方案。移动通信模块150可以包括至少一个滤波器，开关，功率放大器，低噪声放大器(low noise amplifier，LNA)等。移动通信模块150可以由天线1接收电磁波，并对接收的电磁波进行滤波，放大等处理，传送至调制解调处理器进行解调。移动通信模块150还可以对经调制解调处理器调制后的信号放大，经天线1转为电磁波辐射出去。在一些实施例中，移动通信模块150的至少部分功能模块可以被设置于处理器110中。在一些实施例中，移动通信模块150的至少部分功能模块可以与处理器110的至少部分模块被设置在同一个器件中。The mobile communication module 150 can provide solutions for wireless communications including 2G/3G/4G/5G applied to the terminal device 100. The mobile communication module 150 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves from the antenna 1, and filter, amplify, and process the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna 1. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the processor 110. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the same device as at least some of the modules of the processor 110.
调制解调处理器可以包括调制器和解调器。其中，调制器用于将待发送的低频基带信号调制成中高频信号。解调器用于将接收的电磁波信号解调为低频基带信号。随后解调器将解调得到的低频基带信号传送至基带处理器处理。低频基带信号经基带处理器处理后，被传递给应用处理器。应用处理器通过音频设备(不限于扬声器170A，受话器170B等)输出声音信号，或通过显示屏194显示图像或视频。在一些实施例中，调制解调处理器可以是独立的器件。在另一些实施例中，调制解调处理器可以独立于处理器110，与移动通信模块150或其他功能模块设置在同一个器件中。The modem processor may include a modulator and a demodulator. Among them, the modulator is used to modulate the low-frequency baseband signal to be sent into a medium-high frequency signal. The demodulator is used to demodulate the received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After the low-frequency baseband signal is processed by the baseband processor, it is passed to the application processor. The application processor outputs a sound signal through an audio device (not limited to a speaker 170A, a receiver 170B, etc.), or displays an image or video through a display screen 194. In some embodiments, the modem processor may be an independent device. In other embodiments, the modem processor may be independent of the processor 110 and be set in the same device as the mobile communication module 150 or other functional modules.
无线通信模块160可以提供应用在终端设备100上的包括无线局域网(wirelesslocal area networks，WLAN)(如无线保真(wireless fidelity，Wi-Fi)网络)，蓝牙(bluetooth，BT)，全球导航卫星系统(global navigation satellite system，GNSS)，调频(frequency modulation，FM)，近距离无线通信技术(near field communication，NFC)，红外技术(infrared，IR)等无线通信的解决方案。无线通信模块160可以是集成至少一个通信处理模块的一个或多个器件。无线通信模块160经由天线2接收电磁波，将电磁波信号调频以及滤波处理，将处理后的信号发送到处理器110。无线通信模块160还可以从处理器110接收待发送的信号，对其进行调频，放大，经天线2转为电磁波辐射出去。The wireless communication module 160 can provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc., which are applied to the terminal device 100. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves via the antenna 2, modulates the frequency of the electromagnetic wave signal and performs filtering, and sends the processed signal to the processor 110. The wireless communication module 160 can also receive the signal to be sent from the processor 110, modulate the frequency of it, amplify it, and convert it into electromagnetic waves for radiation through the antenna 2.
在一些实施例中，终端设备100的天线1和移动通信模块150耦合，天线2和无线通信模块160耦合，使得终端设备100可以通过无线通信技术与网络以及其他设备通信。所述无线通信技术可以包括全球移动通讯系统(global system for mobile communications，GSM)，通用分组无线服务(general packet radio service，GPRS)，码分多址接入(codedivision multiple access，CDMA)，宽带码分多址(wideband code division multipleaccess，WCDMA)，时分码分多址(time-division code division multiple access，TD-SCDMA)，长期演进(long term evolution，LTE)，BT，GNSS，WLAN，NFC，FM，和/或IR技术等。所述GNSS可以包括全球卫星定位系统(global positioning system，GPS)，全球导航卫星系统(global navigation satellite system，GLONASS)，北斗卫星导航系统(beidounavigation satellite system，BDS)，准天顶卫星系统(quasi-zenith satellitesystem，QZSS)和/或星基增强系统(satellite based augmentation systems，SBAS)。In some embodiments, the antenna 1 of the terminal device 100 is coupled to the mobile communication module 150, and the antenna 2 is coupled to the wireless communication module 160, so that the terminal device 100 can communicate with the network and other devices through wireless communication technology. The wireless communication technology may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), time-division code division multiple access (TD-SCDMA), long term evolution (LTE), BT, GNSS, WLAN, NFC, FM, and/or IR technology. The GNSS may include a global positioning system (GPS), a global navigation satellite system (GLONASS), a Beidou navigation satellite system (BDS), a quasi-zenith satellite system (QZSS) and/or a satellite based augmentation system (SBAS).
终端设备100通过GPU，显示屏194，以及应用处理器等实现显示功能。显示屏194用于显示图像，视频等。The terminal device 100 implements the display function through the GPU, the display screen 194, and the application processor, etc. The display screen 194 is used to display images, videos, etc.
终端设备100可以通过ISP，摄像头193，视频编解码器，GPU，显示屏194以及应用处理器等实现拍摄功能。The terminal device 100 can realize the shooting function through ISP, camera 193, video codec, GPU, display screen 194 and application processor.
数字信号处理器用于处理数字信号，除了可以处理数字图像信号，还可以处理其他数字信号。例如，当终端设备100在频点选择时，数字信号处理器用于对频点能量进行傅里叶变换等。视频编解码器用于对数字视频压缩或解压缩。NPU为神经网络(neural-network，NN)计算处理器，通过借鉴生物神经网络结构，例如借鉴人脑神经元之间传递模式，对输入信息快速处理，还可以不断的自学习。The digital signal processor is used to process digital signals. In addition to processing digital image signals, it can also process other digital signals. For example, when the terminal device 100 selects a frequency point, the digital signal processor is used to perform Fourier transform on the frequency point energy. The video codec is used to compress or decompress digital video. The NPU is a neural network (NN) computing processor that quickly processes input information by drawing on the structure of biological neural networks, such as the transmission mode between neurons in the human brain, and can also continuously self-learn.
外部存储器接口120可以用于连接外部存储卡，例如Micro SD卡，实现扩展终端设备100的存储能力。外部存储卡通过外部存储器接口120与处理器110通信，实现数据存储功能。例如将音乐，视频等文件保存在外部存储卡中。内部存储器121可以用于存储计算机可执行程序代码，所述可执行程序代码包括指令。The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the terminal device 100. The external memory card communicates with the processor 110 through the external memory interface 120 to implement a data storage function. For example, files such as music and videos are stored in the external memory card. The internal memory 121 can be used to store computer executable program codes, which include instructions.
终端设备100可以通过音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，以及应用处理器等实现音频功能。例如音乐播放，录音等。The terminal device 100 can implement audio functions such as music playing and recording through the audio module 170, the speaker 170A, the receiver 170B, the microphone 170C, the earphone interface 170D, and the application processor.
压力传感器180A用于感受压力信号，可以将压力信号转换成电信号。陀螺仪传感器180B可以用于确定终端设备100的运动姿态。磁传感器180D包括霍尔传感器。终端设备100可以利用磁传感器180D检测翻盖皮套的开合。加速度传感器180E可检测终端设备100在各个方向上(一般为三轴)加速度的大小。当终端设备100静止时可检测出重力的大小及方向。还可以用于识别终端姿态，应用于横竖屏切换，计步器等应用。接近光传感器180G可以包括例如发光二极管(LED)和光检测器，例如光电二极管。发光二极管可以是红外发光二极管。终端设备100通过发光二极管向外发射红外光。环境光传感器180L用于感知环境光亮度。终端设备100可以根据感知的环境光亮度自适应调节显示屏194亮度。指纹传感器180H用于采集指纹。温度传感器180J用于检测温度。触摸传感器180K，也称“触控面板”。触摸传感器180K可以设置于显示屏194，由触摸传感器180K与显示屏194组成触摸屏，也称“触控屏”。触摸传感器180K用于检测作用于其上或附近的触摸操作。骨传导传感器180M可以获取振动信号。The pressure sensor 180A is used to sense the pressure signal and can convert the pressure signal into an electrical signal. The gyroscope sensor 180B can be used to determine the motion posture of the terminal device 100. The magnetic sensor 180D includes a Hall sensor. The terminal device 100 can use the magnetic sensor 180D to detect the opening and closing of the flip leather case. The acceleration sensor 180E can detect the magnitude of the acceleration of the terminal device 100 in various directions (generally three axes). When the terminal device 100 is stationary, the magnitude and direction of gravity can be detected. It can also be used to identify the terminal posture and applied to horizontal and vertical screen switching, pedometers and other applications. The proximity light sensor 180G may include, for example, a light emitting diode (LED) and a light detector, such as a photodiode. The light emitting diode may be an infrared light emitting diode. The terminal device 100 emits infrared light outward through the light emitting diode. The ambient light sensor 180L is used to sense the ambient light brightness. The terminal device 100 can adaptively adjust the brightness of the display screen 194 according to the perceived ambient light brightness. The fingerprint sensor 180H is used to collect fingerprints. The temperature sensor 180J is used to detect temperature. The touch sensor 180K is also called a "touch panel". The touch sensor 180K can be set on the display screen 194, and the touch sensor 180K and the display screen 194 form a touch screen, also called a "touch screen". The touch sensor 180K is used to detect touch operations acting on or near it. The bone conduction sensor 180M can obtain vibration signals.
此外，终端设备100还包括气压传感器180C和距离传感器180F。其中，气压传感器180C用于测量气压。在一些实施例中，终端设备100通过气压传感器180C测得的气压值计算海拔高度，辅助定位和导航。In addition, the terminal device 100 also includes an air pressure sensor 180C and a distance sensor 180F. The air pressure sensor 180C is used to measure air pressure. In some embodiments, the terminal device 100 calculates the altitude through the air pressure value measured by the air pressure sensor 180C to assist positioning and navigation.
距离传感器180F，用于测量距离。终端设备100可以通过红外或激光测量距离。在一些实施例中，拍摄场景，终端设备100可以利用距离传感器180F测距以实现快速对焦。The distance sensor 180F is used to measure the distance. The terminal device 100 can measure the distance by infrared or laser. In some embodiments, when shooting a scene, the terminal device 100 can use the distance sensor 180F to measure the distance to achieve fast focusing.
示例性的，终端设备100的软件系统可以采用分层架构，事件驱动架构，微核架构，微服务架构，或云架构。本申请实施例以分层架构的Android系统为例，示例性说明终端设备100的软件结构。图5是本申请实施例的终端设备100的软件结构框图。Exemplarily, the software system of the terminal device 100 can adopt a layered architecture, an event-driven architecture, a micro-core architecture, a micro-service architecture, or a cloud architecture. The present application embodiment takes the Android system of the layered architecture as an example to illustrate the software structure of the terminal device 100. FIG. 5 is a block diagram of the software structure of the terminal device 100 of the present application embodiment.
分层架构将软件分成若干个层，每一层都有清晰的角色和分工。层与层之间通过软件接口通信。在一些实施例中，将Android系统分为四层，从上至下分别为应用程序层，应用程序框架层，安卓运行时(Android runtime)，系统库，内核层，硬件抽象层(hardwareabstraction layer，HAL)以及硬件层。The layered architecture divides the software into several layers, each with clear roles and division of labor. The layers communicate with each other through software interfaces. In some embodiments, the Android system is divided into four layers, from top to bottom: application layer, application framework layer, Android runtime, system library, kernel layer, hardware abstraction layer (HAL), and hardware layer.
应用程序层可以包括一系列应用程序包。如图5所示，应用程序包可以包括相机，日历，地图，WLAN，音乐，短信息，图库，通话，导航，蓝牙，视频等应用程序。The application layer may include a series of application packages. As shown in FIG5 , the application package may include applications such as camera, calendar, map, WLAN, music, short message, gallery, call, navigation, Bluetooth, video, etc.
应用程序框架层为应用程序层的应用程序提供应用编程接口(applicationprogramming interface，API)和编程框架。应用程序框架层包括一些预先定义的函数。如图5所示，应用程序框架层可以包括窗口管理器，内容提供器，电话管理器，资源管理器，通知管理器，视图系统等。The application framework layer provides an application programming interface (API) and a programming framework for the applications in the application layer. The application framework layer includes some predefined functions. As shown in FIG5 , the application framework layer may include a window manager, a content provider, a phone manager, a resource manager, a notification manager, a view system, etc.
窗口管理器用于管理窗口程序。窗口管理器可以获取显示屏大小，判断是否有状态栏，锁定屏幕，截取屏幕等。The window manager is used to manage window programs. The window manager can obtain the display screen size, determine whether there is a status bar, lock the screen, capture the screen, etc.
内容提供器用来存放和获取数据，并使这些数据可以被应用程序访问。所述数据可以包括视频，图像，音频，拨打和接听的电话，浏览历史和书签，电话簿等。Content providers are used to store and retrieve data and make it accessible to applications. The data may include videos, images, audio, calls made and received, browsing history and bookmarks, phone books, etc.
视图系统包括可视控件，例如显示文字的控件，显示图片的控件等。视图系统可用于构建应用程序。显示界面可以由一个或多个视图组成的。例如，包括短信通知图标的显示界面，可以包括显示文字的视图以及显示图片的视图。The view system includes visual controls, such as controls for displaying text, controls for displaying images, etc. The view system can be used to build applications. A display interface can be composed of one or more views. For example, a display interface including a text notification icon can include a view for displaying text and a view for displaying images.
电话管理器用于提供终端设备100的通信功能。例如通话状态的管理(包括接通，挂断等)。The phone manager is used to provide communication functions of the terminal device 100, such as management of call status (including connection, disconnection, etc.).
资源管理器为应用程序提供各种资源，比如本地化字符串，图标，图片，布局文件，视频文件等等。The resource manager provides various resources for applications, such as localized strings, icons, images, layout files, video files, and so on.
通知管理器使应用程序可以在状态栏中显示通知信息，可以用于传达告知类型的消息，可以短暂停留后自动消失，无需用户交互。比如通知管理器被用于告知下载完成，消息提醒等。通知管理器还可以是以图表或者滚动条文本形式出现在系统顶部状态栏的通知，例如后台运行的应用程序的通知，还可以是以对话窗口形式出现在屏幕上的通知。例如在状态栏提示文本信息，发出提示音，终端振动，指示灯闪烁等。The notification manager enables applications to display notification information in the status bar. It can be used to convey notification-type messages and can disappear automatically after a short stay without user interaction. For example, the notification manager is used to notify download completion, message reminders, etc. The notification manager can also be a notification that appears in the system top status bar in the form of a chart or scroll bar text, such as notifications of applications running in the background, or a notification that appears on the screen in the form of a dialog window. For example, a text message is displayed in the status bar, a prompt sound is emitted, the terminal vibrates, the indicator light flashes, etc.
Android Runtime包括核心库和虚拟机。Android runtime负责安卓系统的调度和管理。Android Runtime includes core libraries and virtual machines. Android runtime is responsible for scheduling and management of the Android system.
核心库包含两部分：一部分是java语言需要调用的功能函数，另一部分是安卓的核心库。The core library consists of two parts: one part is the function that needs to be called by the Java language, and the other part is the Android core library.
应用程序层和应用程序框架层运行在虚拟机中。虚拟机将应用程序层和应用程序框架层的java文件执行为二进制文件。虚拟机用于执行障碍物生命周期的管理，堆栈管理，线程管理，安全和异常的管理，以及垃圾回收等功能。The application layer and the application framework layer run in the virtual machine. The virtual machine executes the java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform functions such as obstacle life cycle management, stack management, thread management, security and exception management, and garbage collection.
系统库可以包括多个功能模块。例如：表面管理器(surface manager)，媒体库(Media Libraries)，三维图形处理库(例如：OpenGL ES)，2D图形引擎(例如：SGL)等。The system library may include multiple functional modules, such as surface manager, media library, 3D graphics processing library (such as OpenGL ES), 2D graphics engine (such as SGL), etc.
表面管理器用于对显示子系统进行管理，并且为多个应用程序提供了2D和3D图层的融合。The surface manager is used to manage the display subsystem and provide the fusion of 2D and 3D layers for multiple applications.
媒体库支持多种常用的音频，视频格式回放和录制，以及静态图像文件等。媒体库可以支持多种音视频编码格式，例如:MPEG4，H.264，MP3，AAC，AMR，JPG，PNG等。The media library supports playback and recording of a variety of commonly used audio and video formats, as well as static image files, etc. The media library can support a variety of audio and video encoding formats, such as: MPEG4, H.264, MP3, AAC, AMR, JPG, PNG, etc.
三维图形处理库用于实现三维图形绘图，图像渲染，合成，和图层处理等。The 3D graphics processing library is used to implement 3D graphics drawing, image rendering, compositing, and layer processing.
2D图形引擎是2D绘图的绘图引擎。A 2D graphics engine is a drawing engine for 2D drawings.
内核层是硬件和软件之间的层。内核层至少包含显示驱动，摄像传感器驱动，音频驱动，传感器驱动。The kernel layer is the layer between hardware and software. The kernel layer contains at least display driver, camera sensor driver, audio driver, and sensor driver.
示例性的，如图6所示，为本申请实施例提供的一种云端200的结构示意图。Exemplarily, as shown in FIG6 , it is a schematic diagram of the structure of a cloud 200 provided in an embodiment of the present application.
在一些实施例中，该云端200可以包括获取模块、数据库模块、分类模块、共视判断模块、地图更新模块以及条件判断模块。其中，条件判断模块还可以包括数量统计子模块、时间统计子模块。可选地，该条件判断模块还可以包括语义判断子模块。In some embodiments, the cloud 200 may include an acquisition module, a database module, a classification module, a common view judgment module, a map update module, and a condition judgment module. The condition judgment module may also include a quantity statistics submodule and a time statistics submodule. Optionally, the condition judgment module may also include a semantic judgment submodule.
在一些实施例中，获取模块可以对应于云端200的无线通信模块，可以用于与终端设备100进行无线通信。具体地，获取模块可以通过云端200与终端设备100之间的通信链路接收终端设备100上传的图像，以及图像对应的时间信息、位置信息等。In some embodiments, the acquisition module may correspond to a wireless communication module of the cloud 200, and may be used to wirelessly communicate with the terminal device 100. Specifically, the acquisition module may receive the image uploaded by the terminal device 100, as well as the time information, location information, etc. corresponding to the image, through the communication link between the cloud 200 and the terminal device 100.
在一些实施例中，数据库模块可以用于存储云端获取的已建图区域的离线三维模型。此外，数据库模块还可以用于存储终端设备100在一定历史时长内上传的图像，比如可以存储一个统计周期的图像信息，如一周、一个月、一个季度等。可选地，数据库模块可以位于云端200本地(例如云端200的本地的内存空间)，也可以是云端200的外部存储空间，本申请实施例对此不作限定。In some embodiments, the database module can be used to store the offline three-dimensional model of the mapped area obtained by the cloud. In addition, the database module can also be used to store images uploaded by the terminal device 100 within a certain historical period, such as image information of a statistical period, such as a week, a month, a quarter, etc. Optionally, the database module can be located locally in the cloud 200 (for example, the local memory space of the cloud 200), or it can be an external storage space of the cloud 200, which is not limited in the embodiments of the present application.
可选地，云端200可以对数据库中的无效图像信息进行清理。这里的无效图像信息可以包括点云地图更新之后，之前统计周期内收到的图像信息，也即不再用于地图更新流程的陈旧图像信息。具体地，云端清除陈旧图像信息的操作可以由图像清理模块(图中未示出)执行。Optionally, the cloud 200 can clean up invalid image information in the database. The invalid image information here may include image information received in the previous statistical period after the point cloud map is updated, that is, old image information that is no longer used in the map update process. Specifically, the operation of clearing out old image information in the cloud can be performed by an image cleaning module (not shown in the figure).
在一些实施例中，分类模块可以用于将获取到query图像按照位置信息进行分类，或者按照位置信息和时间信息进行分类。以按照位置信息对query图像进行分类为例，如图7所示，本申请实施例提供的第一GPS区域对应的位置示意图。结合图示，可以获知该第一GPS区域是以(X11,Y11)、(X12,Y12)、(X13,Y13)、(X14,Y14)四个坐标的位置作为顶点的矩形区域，并且该区域包括坐标为(xa,ya)的位置，以及坐标为(xb,yb)的位置。假如云端获取了两张query图像，其位置信息分别为(xa,ya)和(xb,yb)，那么云端就可以根据query图像的坐标和第一GPS区域的位置，判断这两张query图像同属于第一GPS区域，之后将这两张query图像信息划分到第一GPS区域对应的类别中。In some embodiments, the classification module can be used to classify the query images acquired according to the location information, or according to the location information and time information. Taking the classification of query images according to the location information as an example, as shown in FIG7, a schematic diagram of the location corresponding to the first GPS area provided in an embodiment of the present application. In combination with the diagram, it can be known that the first GPS area is a rectangular area with the positions of the four coordinates (X11 ,Y11 ), (X12 ,Y12 ), (X13 ,Y13 ), and (X14 ,Y14 ) as vertices, and the area includes a position with coordinates (xa ,ya ) and a position with coordinates (xb ,yb ). If the cloud obtains two query images, and their location information is (xa ,ya ) and (xb ,yb ), respectively, then the cloud can determine that the two query images belong to the first GPS area based on the coordinates of the query images and the location of the first GPS area, and then classify the two query image information into the category corresponding to the first GPS area.
在一些实施例中，针对已建图区域划分得到的GPS区域可以是任意形状，比如矩形(如图2或图7所示)或者圆形等。为了便于理解，请参见表1，本申请实施例以GPS区域的形状是矩形为例，提供了一种按照位置信息对query图像进行划分的分类结果示例。In some embodiments, the GPS area obtained by dividing the mapped area can be of any shape, such as a rectangle (as shown in FIG. 2 or FIG. 7 ) or a circle, etc. For ease of understanding, please refer to Table 1. The embodiment of the present application takes the shape of the GPS area as a rectangle as an example, and provides an example of classification results of dividing the query image according to location information.
表1Table 1
需要说明的是，上述表1所示的分类结果仅为示例。在实际应用中，该分类结果还可以包括其他更多的信息项，比如图像对应的时间信息、图像所对应的季节(可参见下表2)等，本申请实施例对此不作限定。It should be noted that the classification results shown in Table 1 above are only examples. In practical applications, the classification results may also include other more information items, such as the time information corresponding to the image, the season corresponding to the image (see Table 2 below), etc., which are not limited in this embodiment of the present application.
在一些实施例中，条件判断模块可以基于云端200获取的query图像判断对应的GPS区域是否满足更新点云地图的第一条件，也即首先判断该GPS区域是否需要更新点云地图。若需要更新点云地图，则可以对该GPS区域执行后续的点云地图更新流程。示例性的，条件判断模块可以具体包括数量统计子模块、时间统计子模块和语义判断子模块。其中，数量统计子模块可以用于统计异常定位图像的数量，或者可以用于统计异常定位图像的数量和正常定位图像的数量。时间统计子模块可以用于统计异常定位图像的占比随时间的变化情况。语义判断子模块可以用于对同一GPS区域内的正常定位图像和异常定位图像进行语义分割，并统计显著语义类别，判断两种类型图像的语义类别的差距。In some embodiments, the conditional judgment module can judge whether the corresponding GPS area meets the first condition for updating the point cloud map based on the query image obtained by the cloud 200, that is, first judge whether the point cloud map of the GPS area needs to be updated. If the point cloud map needs to be updated, the subsequent point cloud map update process can be executed for the GPS area. Exemplarily, the conditional judgment module can specifically include a quantity statistics submodule, a time statistics submodule and a semantic judgment submodule. Among them, the quantity statistics submodule can be used to count the number of abnormal positioning images, or can be used to count the number of abnormal positioning images and the number of normal positioning images. The time statistics submodule can be used to count the changes in the proportion of abnormal positioning images over time. The semantic judgment submodule can be used to perform semantic segmentation on normal positioning images and abnormal positioning images in the same GPS area, and count significant semantic categories to determine the gap between the semantic categories of the two types of images.
在一些实施例中，可以根据数量统计子模块和时间统计子模块的输出结果判断是否更新点云地图；或者，根据数量统计子模块、时间统计子模块以及语义判断子模块的输出结果判断是否更新点云地图。In some embodiments, whether to update the point cloud map can be determined based on the output results of the quantity statistics submodule and the time statistics submodule; or, whether to update the point cloud map can be determined based on the output results of the quantity statistics submodule, the time statistics submodule and the semantic judgment submodule.
在一些实施例中，共视判断模块可以用于对GPS区域所有的query图像(包括正常定位图像和异常定位图像)进行特征匹配，构建场景图(scene graph)。此外，共视判断模块还可以用于判断场景图的连通性以及正常定位图像的数量，以保证需要更新的地图部分能够更好地融合到原点云地图中。其中，具体的处理过程及相关原理将在下文实施例中进行说明，此处暂不详述。In some embodiments, the common view judgment module can be used to perform feature matching on all query images (including normal positioning images and abnormal positioning images) in the GPS area to construct a scene graph. In addition, the common view judgment module can also be used to determine the connectivity of the scene graph and the number of normal positioning images to ensure that the map portion that needs to be updated can be better integrated into the origin cloud map. Among them, the specific processing process and related principles will be described in the following embodiments and will not be described in detail here.
需要说明的是，图6所示的云端200的结构仅为示例。在实际应用中，云端200还可以包括其他更多或更少的模块，或者云端200包括的模块还可以用于执行其他更多的功能，本申请实施例对此不作限定。It should be noted that the structure of the cloud 200 shown in Figure 6 is only an example. In practical applications, the cloud 200 may also include more or fewer modules, or the modules included in the cloud 200 may also be used to perform more functions, which is not limited in the embodiments of the present application.
以下结合附图，对本申请提供的点云地图更新的方法可能涉及的实现场景进行介绍。示例性的，如图8A至图8D所示，为本申请实施例提供的地图更新的方法在实现过程可能涉及到的一些场景示意图。The following is an introduction to the implementation scenarios that may be involved in the method for updating the point cloud map provided in this application in conjunction with the accompanying drawings. For example, as shown in Figures 8A to 8D, some schematic diagrams of scenarios that may be involved in the implementation process of the method for updating the map provided in the embodiment of this application.
首先请参见图8A，为第一GPS区域示意图。在一些场景中，当用户进入该第一GPS区域后，终端设备(下文以手机为例)可以开启VPS定位功能。示例性的，VPS定位功能开启的方式可以有多种，比如：方式1，用户在进入该GPS区域之后，可以手动开启手机自带的VPS定位功能；方式2，在获得用户授权手机使用VPS功能的基础上，当检测到用户进入第一GPS区域时(如基于手机的GPS位置)，手机可以自动开启VPS定位功能。First, please refer to Figure 8A, which is a schematic diagram of the first GPS area. In some scenarios, when the user enters the first GPS area, the terminal device (hereinafter taking the mobile phone as an example) can turn on the VPS positioning function. Exemplarily, there are many ways to turn on the VPS positioning function, such as: Method 1, after entering the GPS area, the user can manually turn on the VPS positioning function of the mobile phone; Method 2, based on obtaining the user's authorization to use the VPS function, when the user enters the first GPS area (such as based on the GPS location of the mobile phone), the mobile phone can automatically turn on the VPS positioning function.
在一些实施例中，VPS定位功能可以是手机出厂自带的功能，也可以是通过第三方应用程序(application，App)实现的功能。一个手机可以从不同角度拍摄同一GPS区域的多张图像，如图8A所示，手机可以从不同角度拍摄的同一建筑物，如图示的第一角度、第二角度和第三角度。不同角度的图像可以呈现该建筑物上不同部分的特征。比如，第一角度对应的图像可以如图8B所示；第二角度对应的图像可以如图8C所示；第三角度对应的图像可以如图8D所示。In some embodiments, the VPS positioning function can be a function that comes with the mobile phone when it leaves the factory, or it can be a function implemented by a third-party application (application, App). A mobile phone can take multiple images of the same GPS area from different angles. As shown in FIG8A , the mobile phone can take pictures of the same building from different angles, such as the first angle, the second angle, and the third angle shown in the figure. Images at different angles can present the features of different parts of the building. For example, the image corresponding to the first angle can be shown in FIG8B ; the image corresponding to the second angle can be shown in FIG8C ; the image corresponding to the third angle can be shown in FIG8D .
之后，手机将拍摄的图像上传至云端，随图像一起上传的还包括图像对应的位置信息和时间信息等。在一些实施例中，手机向云端上传图像信息的方式可以有多种，比如：方式(1)，手机可以实时地向云端上传图像信息；方式(2)，手机可以在用户的手动操作下向云端上传图像信息；方式(3)，手机可以周期性地向云端上传图像信息(如每天固定时刻上传)；方式(4)，手机可以在连接到特定类型的网络时，向云端上传图像信息，如无线保真(wireless fidelity，Wi-Fi)网络。本申请实施例对手机向云端上传图像信息的具体方式不做限定。Afterwards, the mobile phone uploads the captured image to the cloud, and the location information and time information corresponding to the image are also uploaded together with the image. In some embodiments, there are multiple ways for the mobile phone to upload image information to the cloud, such as: method (1), the mobile phone can upload image information to the cloud in real time; method (2), the mobile phone can upload image information to the cloud under the manual operation of the user; method (3), the mobile phone can upload image information to the cloud periodically (such as uploading at a fixed time every day); method (4), the mobile phone can upload image information to the cloud when connected to a specific type of network, such as a wireless fidelity (Wi-Fi) network. The embodiments of the present application do not limit the specific method for the mobile phone to upload image information to the cloud.
在一些实施例中，云端获取图像之后，可以按照位置信息对上传的图像进行分类。之后，云端通过VPS对同一GPS区域的图像定位，获取该GPS区域中的正常定位图像和异常定位图像。再之后，云端可以根据第一条件，判断是否触发更新点云地图的后续流程。若执行后续流程，则云端可以通过共视判断，构造场景图，并根据第二条件判断更新部分可否较好地融入原有的离线三维模型/原有的点云地图。若满足第二条件，则云端可以对该GPS区域全景图中的两种类型图像的特征点进行三维转化，获取该部分对应的三维点云数据，然后利用这些三维点云数据更新原有离线三维模型中的相应部分。该第二条件可以用于判断场景图的连通性、以及与原有点云模型其他部分的融合效果。In some embodiments, after the cloud acquires the image, the uploaded image can be classified according to the location information. Afterwards, the cloud locates the image in the same GPS area through VPS, and obtains the normal positioning image and the abnormal positioning image in the GPS area. After that, the cloud can determine whether to trigger the subsequent process of updating the point cloud map according to the first condition. If the subsequent process is executed, the cloud can construct a scene graph through common vision judgment, and determine whether the updated part can be better integrated into the original offline three-dimensional model/original point cloud map according to the second condition. If the second condition is met, the cloud can perform three-dimensional transformation on the feature points of the two types of images in the panoramic view of the GPS area, obtain the three-dimensional point cloud data corresponding to the part, and then use these three-dimensional point cloud data to update the corresponding part in the original offline three-dimensional model. The second condition can be used to judge the connectivity of the scene graph and the fusion effect with other parts of the original point cloud model.
需要说明的是，云端通过不同角度的图像，提取建筑物在不同角度下的特征，可以便于云端获取建筑物较全面的形貌，还可以增加图像特征与三维模型匹配的准确性。换句话说，点云地图更新的准确性与云端获取的图像的数量以及角度的多样性关联。如果云端获取某一GPS区域的图像的数量越多，且角度越多样化，则云端针对该GPS区域的地图更新的准确性可能更高。It should be noted that the cloud can extract the features of buildings at different angles through images from different angles, which can facilitate the cloud to obtain a more comprehensive shape of the building and increase the accuracy of matching image features with three-dimensional models. In other words, the accuracy of point cloud map updates is related to the number of images obtained by the cloud and the diversity of angles. If the cloud obtains more images of a certain GPS area and the more diverse the angles, the accuracy of the cloud map update for that GPS area may be higher.
根据本申请实施例提供的点云地图更新的方法，可以通过云端与终端设备的交互自动获取已建图区域的图像信息，并根据获取的图像信息在准确的时机触发对点云地图的更新操作。整个点云地图更新过程无需过度耗费人力，不仅能够极大地节约地图更新的成本，还能够更加智能化地提升地图更新时机的准确性，保持了地图鲜度。According to the method for updating the point cloud map provided in the embodiment of the present application, the image information of the mapped area can be automatically obtained through the interaction between the cloud and the terminal device, and the update operation of the point cloud map can be triggered at the right time according to the obtained image information. The entire point cloud map update process does not require excessive manpower, which can not only greatly save the cost of map updates, but also more intelligently improve the accuracy of map update timing and maintain the freshness of the map.
上文对本申请实施例提供的点云地图更新的方法中可能涉及到的设备的结构、系统架构、应用场景以及可视层面的实现流程进行了介绍。为了更好地理解本申请实施例提供的方法，以下结合附图对该方法的底层实现逻辑进行介绍。The above describes the structure of the device, system architecture, application scenarios, and visual implementation process that may be involved in the method for updating the point cloud map provided in the embodiment of the present application. In order to better understand the method provided in the embodiment of the present application, the underlying implementation logic of the method is introduced below in conjunction with the accompanying drawings.
示例性的，如图9所示，为本申请实施例提供的一种点云地图更新的方法的示意性流程图。该流程可以由终端设备(仍以手机为例)和云端作为主体来执行，具体可以包括以下步骤：As shown in FIG9 , it is a schematic flow chart of a method for updating a point cloud map provided in an embodiment of the present application. The process can be performed by a terminal device (still taking a mobile phone as an example) and a cloud as the main body, and can specifically include the following steps:
S901，用户在已建图区域使用VPS定位功能。S901, the user uses the VPS positioning function in the mapped area.
其中，这里的已建图区域可以指预先建立有离线三维模型的区域，比如该已建图区域是一个商场时，说明云端已存储有该商场对应的离线三维模型，也即点云地图。The mapped area here may refer to an area for which an offline 3D model has been pre-established. For example, when the mapped area is a shopping mall, it means that the cloud has stored an offline 3D model corresponding to the shopping mall, that is, a point cloud map.
在一些实施例中，用户可以在已建图区域主动使用VPS定位功能。例如，某些已建图区域的管理后台可以通过推送、通知等方式告知用户在该区域内可以使用VPS定位功能，基于此，用户可以主动开启该功能，或者授权手机使用该功能。In some embodiments, the user can actively use the VPS positioning function in the mapped area. For example, the management backend of some mapped areas can inform the user that the VPS positioning function can be used in the area through push notifications, etc. Based on this, the user can actively turn on the function or authorize the mobile phone to use the function.
在一种可能的情况下，手机还可以向云端上报用户的GPS位置，当云端检测到用户进入到已建图区域的GPS范围内，可以指示手机开启VPS定位功能，或者通过手机提示用户同意开启VPS定位功能，或者通过手机提示用户手动开启VPS定位功能。之后，手机可以自动地或者经由用户授权使用或者用户手动操作后，开启VPS定位功能。In one possible scenario, the mobile phone can also report the user's GPS location to the cloud. When the cloud detects that the user has entered the GPS range of the mapped area, it can instruct the mobile phone to turn on the VPS positioning function, or prompt the user to agree to turn on the VPS positioning function through the mobile phone, or prompt the user to manually turn on the VPS positioning function through the mobile phone. After that, the mobile phone can turn on the VPS positioning function automatically, or after the user's authorization or manual operation.
在一些实施例中，手机开启VPS定位功能之后，可以自动采集环境中的图像，或者在用户手动操作下采集环境中的图像。可选地，当自动采集图像时，手机可以进行周期性采集，例如3秒，10秒等。In some embodiments, after the mobile phone turns on the VPS positioning function, it can automatically collect images in the environment, or collect images in the environment under manual operation of the user. Optionally, when automatically collecting images, the mobile phone can perform periodic collection, such as 3 seconds, 10 seconds, etc.
S902，手机向云端上传图像，云端保存所有使用VPS定位的query图像，及其时间信息、位置信息。S902, the mobile phone uploads the image to the cloud, and the cloud saves all query images located using the VPS, as well as their time information and location information.
其中，query图像是指云端获取的，由手机上传的某GPS区域的图像。在一些描述中，结合上下文意思，query图像也被直接描述为图像。在实际应用中，该query图像可以是RGB图像或者图片序列。The query image refers to an image of a certain GPS area obtained from the cloud and uploaded by a mobile phone. In some descriptions, the query image is also directly described as an image in combination with the context. In actual applications, the query image can be an RGB image or a picture sequence.
在一些实施例中，手机在向云端发送图像时，还可以向云端发送该图像对应的位置信息(如GPS位置)。此外，还可以向云端发送该图像对应的时间信息，如在图像信息中添加时间戳。可选地，时间信息例如可以是图像被采集时手机的系统时间，或者手机向云端发送该图像时的系统时间。In some embodiments, when the mobile phone sends an image to the cloud, it can also send the location information (such as GPS location) corresponding to the image to the cloud. In addition, the time information corresponding to the image can also be sent to the cloud, such as adding a timestamp to the image information. Optionally, the time information can be, for example, the system time of the mobile phone when the image is captured, or the system time when the mobile phone sends the image to the cloud.
S903，云端对上传的图像按照位置信息进行分类，并判断VPS定位是否成功。S903, the cloud classifies the uploaded images according to the location information and determines whether the VPS positioning is successful.
在一些实施例中，云端可以按照图像对应的GPS位置进行分类。示例性的，该过程例如可以包括：云端预先将大面积的已建图区域划分为多个面积较小的GPS区域；每个GPS区域都可以对应一个尺寸范围值(如10m×10m)和GPS范围值，比如图7所示的矩形GPS区域，其GPS范围值可以根据该矩形的四个顶点确定。之后，云端根据query图像的GPS位置判断query图像属于哪一个GPS区域，并将位于同一个GPS区域的query图像归于同一类。示例性的，分类后的结果例如可以参见上表1所示。In some embodiments, the cloud can classify images according to the GPS location corresponding to the image. Exemplarily, the process may include, for example: the cloud divides a large area of mapped area into multiple smaller GPS areas in advance; each GPS area may correspond to a size range value (such as 10m×10m) and a GPS range value, such as the rectangular GPS area shown in Figure 7, whose GPS range value can be determined based on the four vertices of the rectangle. Afterwards, the cloud determines which GPS area the query image belongs to based on the GPS location of the query image, and classifies query images located in the same GPS area into the same category. Exemplarily, the results of the classification can be shown in Table 1 above.
可选地，GPS区域对应的形状也可以设定为圆形或其它形状。其中，若GPS区域为圆形，该GPS区域的位置例如可以根据圆心的位置和半径确定。Optionally, the shape corresponding to the GPS area can also be set to a circle or other shapes. Wherein, if the GPS area is a circle, the position of the GPS area can be determined according to the position of the center of the circle and the radius, for example.
需要说明的是，由于相机的视野范围有限，每张图像展现的范围有限，因而将整个大环境划分为多个小块的GPS区域，就可以针对这些规模较小的GPS区域，逐个进行点云地图更新处理，或者针对几个GPS区域并行进行点云地图更新处理，而无须对整体已建图区域进行一次性大规模处理，提高了地图更新处理的灵活性以及稳定性。It should be noted that due to the limited field of view of the camera, the display range of each image is limited. Therefore, the entire environment is divided into multiple small GPS areas. Point cloud map update processing can be performed one by one for these smaller GPS areas, or point cloud map update processing can be performed for several GPS areas in parallel without the need for one-time large-scale processing of the entire mapped area, which improves the flexibility and stability of map update processing.
在一些实施例中，云端判断VPS定位是否成功可以依据query图像与离线三维模型中对应图像的图像匹配度来判断。例如，query图像与离线三维模型中对应位置图像的图像匹配度等于或大于某个预设的第一数值(如75％、80％等)，则认为该query图像对应的定位结果准确；或者，query图像与离线三维模型中对应图像的特征匹配数目等于或大于预设的第二数值，则认为该query图像对应的定位结果准确。Query图像与离线三维模型中对应图像的图像匹配度小于预设的第一数值(如75％、80％等)，则认为该query图像对应的定位信息不准确；或者，query图像与离线三维模型中对应图像的特征匹配数目小于预设的第二数值，则认为该query图像对应的定位信息不准确。可选地，云端确定query图像的定位结果准确性之后，可以向手机反馈定位结果信息，提示当前定位准确；或者，提示当前定位不准确。In some embodiments, the cloud determines whether the VPS positioning is successful based on the image matching degree between the query image and the corresponding image in the offline three-dimensional model. For example, if the image matching degree between the query image and the image at the corresponding position in the offline three-dimensional model is equal to or greater than a preset first value (such as 75%, 80%, etc.), the positioning result corresponding to the query image is considered accurate; or, if the number of feature matches between the query image and the corresponding image in the offline three-dimensional model is equal to or greater than a preset second value, the positioning result corresponding to the query image is considered accurate. If the image matching degree between the query image and the corresponding image in the offline three-dimensional model is less than a preset first value (such as 75%, 80%, etc.), the positioning information corresponding to the query image is considered inaccurate; or, if the number of feature matches between the query image and the corresponding image in the offline three-dimensional model is less than a preset second value, the positioning information corresponding to the query image is considered inaccurate. Optionally, after the cloud determines the accuracy of the positioning result of the query image, it can feedback the positioning result information to the mobile phone, prompting that the current positioning is accurate; or prompting that the current positioning is inaccurate.
可以理解的是，如果query图像的定位结果准确，则意味着该区域当前的真实环境与离线三维模型对应的环境差异较小，相比于建模时期，该环境未发生明显变化，故该情形下点云地图需要更新的可能性较小。如果query图像的定位结果不准确，则意味着该区域当前的真实环境与离线三维模型对应的环境差异较大，相比于建模时期，该环境可能发生了明显变化，故该情形下点云地图需要更新的可能性较大。It is understandable that if the positioning result of the query image is accurate, it means that the current real environment of the area is slightly different from the environment corresponding to the offline 3D model, and the environment has not changed significantly compared to the modeling period, so the point cloud map is less likely to need to be updated in this case. If the positioning result of the query image is inaccurate, it means that the current real environment of the area is significantly different from the environment corresponding to the offline 3D model, and the environment may have changed significantly compared to the modeling period, so the point cloud map is more likely to need to be updated in this case.
需要说明的是，在本申请VPS定位过程中用于获取图像匹配度的特征可以是一种图像描述符(image descriptor)。简单地说，图像描述符的输出是特征向量，是图像本身的抽象，它可以是用于表示图像的数字列表。以局部聚合描述符的向量(vector of locallyaggregated descriptors，Vlad)为例，提取图像描述符的过程可以包括：将待提取图像(即query图像)输入卷积神经网络，可以获取H*W*D(H为待提取图像的高度，W为待提取图像的宽度，D为特征维度)的特征矩阵；将该特征矩阵输入卷积网络层，通过卷积和归一化指数函数，可以获取聚类中心，并获取待提取图像的局部特征与聚类中心的残差分布；最后利用该残差分布，通过归一化操作，输出待提取图像的图像描述符。It should be noted that the feature used to obtain the image matching degree in the VPS positioning process of the present application can be an image descriptor. Simply put, the output of the image descriptor is a feature vector, which is an abstraction of the image itself, and it can be a list of numbers used to represent the image. Taking the vector of locally aggregated descriptors (Vlad) as an example, the process of extracting the image descriptor may include: inputting the image to be extracted (i.e., the query image) into the convolutional neural network, and obtaining a feature matrix of H*W*D (H is the height of the image to be extracted, W is the width of the image to be extracted, and D is the feature dimension); inputting the feature matrix into the convolutional network layer, through convolution and normalized exponential function, the cluster center can be obtained, and the residual distribution of the local features of the image to be extracted and the cluster center can be obtained; finally, using the residual distribution, through normalization operation, the image descriptor of the image to be extracted is output.
在一些实施例中，在提取出query图像的图像描述符之后，还可以通过暴力匹配算法或者快速最近邻算法(fast libary for approximate nearest neighbors，FLANN)等进行匹配，获取图像匹配结果。可选地，在前述图像匹配过程中，为了结果更加准确，还可以通过交叉匹配算法等进行匹配结果优化，该具体实现流程可以参见现有技术，本申请对此不作赘述。In some embodiments, after extracting the image descriptor of the query image, it is also possible to perform matching through a brute force matching algorithm or a fast library for approximate nearest neighbors (FLANN) algorithm to obtain an image matching result. Optionally, in the aforementioned image matching process, in order to make the result more accurate, the matching result can also be optimized through a cross matching algorithm, etc. The specific implementation process can be referred to the prior art, and this application will not elaborate on this.
S904，云端进行异常判断，也即判断是否需要对GPS区域的点云地图进行更新。S904, the cloud performs abnormality judgment, that is, judges whether the point cloud map of the GPS area needs to be updated.
在一些实施例中，云端可以根据预设的第一条件判断是否需要更新GPS区域的点云地图。其中，根据第一条件判断主要包括：云端根据异常定位图像的数量和/或异常定位图像的占比随时间的变化，判断是否需要更新GPS区域的点云地图。In some embodiments, the cloud can determine whether the point cloud map of the GPS area needs to be updated according to a preset first condition. The determination according to the first condition mainly includes: the cloud determines whether the point cloud map of the GPS area needs to be updated according to the change in the number of abnormal positioning images and/or the proportion of abnormal positioning images over time.
示例性的，本申请实施例中用于判断是否更新点云地图的第一条件可以根据需要设置为不同内容。比如，该第一条件可以设置为：若GPS区域中的异常定位图像的数量等于或大于第一阈值，则确定需要更新该GPS区域对应的点云地图。再比如，该第一条件还可以设置为：若GPS区域中的异常定位图像的数量等于或大于第一阈值，以及该GPS区域的异常定位图像的占比在某一时间(定义为异常时间)等于或大于第二阈值，则确定需要更新该GPS区域对应的点云地图。又比如，该第一条件还可以设置为：若GPS区域中的异常定位图像的数量等于或大于第一阈值，以及该GPS区域的异常定位图像占比在异常时间等于或大于第二阈值，以及正常定位图像和异常定位图像进行语义分割后，语义类别差距大于预设差距，则确定需要更新该GPS区域对应的点云地图。其中，基于正常定位图像和异常定位图像的语义类别差距结果判断是否更新点云地图的步骤，可以是可选的。Exemplarily, the first condition for determining whether to update the point cloud map in the embodiment of the present application can be set to different contents as needed. For example, the first condition can be set to: if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold, it is determined that the point cloud map corresponding to the GPS area needs to be updated. For another example, the first condition can also be set to: if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold, and the proportion of abnormal positioning images in the GPS area is equal to or greater than the second threshold at a certain time (defined as the abnormal time), it is determined that the point cloud map corresponding to the GPS area needs to be updated. For another example, the first condition can also be set to: if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold, and the proportion of abnormal positioning images in the GPS area is equal to or greater than the second threshold at the abnormal time, and after the normal positioning image and the abnormal positioning image are semantically segmented, the semantic category gap is greater than the preset gap, then it is determined that the point cloud map corresponding to the GPS area needs to be updated. Among them, the step of determining whether to update the point cloud map based on the semantic category gap result of the normal positioning image and the abnormal positioning image can be optional.
其中，这里所说的异常定位图像的占比是指异常定位图像的数量在同一类别的query图像中所占的比例。该同一类别是指云端对query图像分类的类别，比如同一GPS区域、同一季节等。The ratio of abnormal positioning images mentioned here refers to the ratio of the number of abnormal positioning images to the query images of the same category. The same category refers to the category in which the cloud classifies the query images, such as the same GPS area, the same season, etc.
针对上述第一种类型的第一条件，判断的过程可以包括以下步骤：For the first condition of the first type mentioned above, the judging process may include the following steps:
S904(1)a，云端统计其获取的异常定位图像的数量，该统计方式例如可以是周期性的，如每周、每个月或者每个季度等。S904(1)a, the cloud counts the number of abnormal positioning images it obtains. The counting method may be periodic, such as weekly, monthly, or quarterly.
S904(1)b，若在统计周期内，GPS区域的异常定位图像的数量等于或大于第一阈值，则云端执行对地图的更新流程，也即可以执行下文中的步骤S905和步骤S906。S904(1)b, if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold value within the statistical period, the cloud executes the map update process, that is, the following steps S905 and S906 can be executed.
针对上述第二种类型的第一条件，判断的过程可以包括以下步骤：For the first condition of the second type above, the judging process may include the following steps:
S904(2)a，云端统计其获取的异常定位图像的数量，该统计方式例如可以是周期性的，如每周、每个月或者每个季度等。S904(2)a, the cloud counts the number of abnormal positioning images it obtains. The counting method may be periodic, such as weekly, monthly, or quarterly.
S904(2)b，若在统计周期内，GPS区域的异常定位图像的数量等于或大于第一阈值，则统计该周期内，异常定位图像的占比是否在异常时间等于或大于第二阈值。S904(2)b, if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold within the statistical period, then count whether the proportion of abnormal positioning images during the period is equal to or greater than the second threshold at the abnormal time.
需要说明的是，考虑到GPS区域内的行人活动、季节更替等因素会对图像所呈现的真实环境的准确性带来不可避免的影响，如行人遮挡某部分商场布局，季节更替导致树木形态变化等，在正常情况下，异常定位图像的占比通常会固定在某个比例(如在15％)附近。但如果GPS区域发生实质性变化，比如增加或拆减建筑物、道路更改、广告牌更换等，那么异常定位图像的占比就会在某个时间(即异常时间)之后明显增加。因而通过步骤S904(2)b可以解决不可避免因素会对更新点云地图带来误差的问题。It should be noted that, considering the pedestrian activities and seasonal changes in the GPS area, the accuracy of the real environment presented by the image will inevitably be affected. For example, pedestrians block part of the shopping mall layout, seasonal changes cause changes in the shape of trees, etc. Under normal circumstances, the proportion of abnormal positioning images is usually fixed at a certain proportion (such as around 15%). However, if there are substantial changes in the GPS area, such as adding or demolishing buildings, changing roads, replacing billboards, etc., then the proportion of abnormal positioning images will increase significantly after a certain time (i.e., abnormal time). Therefore, step S904 (2) b can solve the problem that inevitable factors will cause errors in updating the point cloud map.
在一些实施例中，考虑到季节变化对真实环境的影响，云端还可以再按照图像的时间戳所在的季节对query图像进一步分类。示例性的，如下表2所示，为本申请实施例提供的另一种query图像分类结果示例。In some embodiments, considering the impact of seasonal changes on the real environment, the cloud can further classify the query image according to the season of the image timestamp. For example, as shown in Table 2 below, another example of query image classification results provided in an embodiment of the present application.
表2Table 2
需要说明的是，在本申请实施例中，按照季节进行图像分类可以作为可选方式。容易理解地，在实际应用中，针对不同地域，可以灵活选择是否按照图像时间进行分类，或者灵活地按其他时间划分方式(如按照每周、每月作为周期)进行分类，而无需必须按照传统的四季划分的形式进行分类。比如，在四季分明的地区(如黄河流域附近区域)，可以按照上述表2的方式进行分类；然而，如果在四季不分明的地区(如赤道附近区域)，则可以无需按照季节对图像进行分类。It should be noted that in the embodiments of the present application, image classification by season can be used as an optional method. It is easy to understand that in practical applications, for different regions, you can flexibly choose whether to classify according to image time, or flexibly classify according to other time division methods (such as weekly or monthly cycles), without having to classify according to the traditional four-season division. For example, in areas with distinct four seasons (such as areas near the Yellow River basin), classification can be performed in accordance with the method in Table 2 above; however, if in areas with unclear four seasons (such as areas near the equator), there is no need to classify images by season.
或者，在室外环境可以按照时间信息进行图像分类，而在室内环境则无需按照时间信息进行图像分类。在一些实施例中，云端可以通过图像识别确定该图像对应的环境是室内还是室外，例如可以通过图像中天空的占比进行识别。Alternatively, in an outdoor environment, images may be classified according to time information, but in an indoor environment, images do not need to be classified according to time information. In some embodiments, the cloud can determine whether the environment corresponding to the image is indoors or outdoors through image recognition, for example, by the proportion of the sky in the image.
S904(2)c，若该周期内，异常定位图像的占比在异常时间等于或大于第二阈值，则云端执行对地图的更新流程，也即可以执行下文中的步骤S905和步骤S906。S904(2)c, if the proportion of abnormal positioning images during the abnormal time is equal to or greater than the second threshold value within the period, the cloud executes the map update process, that is, the following steps S905 and S906 can be executed.
针对上述第三种类型的第一条件，判断的过程可以包括以下步骤：For the first condition of the third type above, the judging process may include the following steps:
S904(3)a，云端统计其获取的异常定位图像的数量。S904(3)a, the cloud counts the number of abnormal positioning images it obtains.
S904(3)b，若在统计周期内，GPS区域的异常定位图像的数量等于或大于第一阈值，则统计该周期内，异常定位图像的占比是否在异常时间等于或大于第二阈值。S904(3)b, if the number of abnormal positioning images in the GPS area is equal to or greater than the first threshold within the statistical period, then count whether the proportion of abnormal positioning images within the period is equal to or greater than the second threshold at the abnormal time.
S904(3)c，若在该周期内，异常定位图像的占比在异常时间等于或大于第二阈值，则统计该GPS区域内异常定位图像和正常定位图像的语义类别的差距是否达到目标差距。S904(3)c, if within the period, the proportion of abnormal positioning images at the abnormal time is equal to or greater than the second threshold, then statistics are performed to see whether the difference in semantic categories between the abnormal positioning images and the normal positioning images in the GPS area reaches the target difference.
在一些实施例中，云端可以对GPS区域内异常定位图像和正常定位图像进行语义分割，并统计显著语义类别，之后判断异常定位图像和正常定位图像的语义类别是否相差目标差距。其中，云端对异常定位图像和正常定位图像进行语义分割的方式可以基于传统方法和/或基于卷积神经网络(convolutional neural networks，CNN)的方法。其中，传统的语义分割方法例如包括基于统计的方法和基于几何的方法；基于CNN的方法例如包括基于语义分割CNN模型的方法，或者基于全卷积网络(fully convolutional networks，FCN)模型的方法等等。语义分割的详细流程可以参见现有流程，本申请实施例对此不再详述。In some embodiments, the cloud can perform semantic segmentation on abnormal positioning images and normal positioning images in the GPS area, and count significant semantic categories, and then determine whether the semantic categories of the abnormal positioning images and the normal positioning images differ by a target difference. Among them, the cloud can perform semantic segmentation on abnormal positioning images and normal positioning images based on traditional methods and/or methods based on convolutional neural networks (CNN). Among them, traditional semantic segmentation methods include, for example, statistical-based methods and geometric-based methods; CNN-based methods include, for example, methods based on semantic segmentation CNN models, or methods based on fully convolutional networks (FCN) models, etc. The detailed process of semantic segmentation can be referred to the existing process, and the embodiments of the present application will not be described in detail.
S904(3)d，若该GPS区域异常定位图像和正常定位图像的语义类别的差距达到目标差距，则云端执行对地图的更新流程，也即可以执行下文中的步骤S905和步骤S906。S904(3)d, if the difference in semantic categories between the abnormal positioning image and the normal positioning image of the GPS area reaches the target difference, the cloud executes the map update process, that is, the following steps S905 and S906 can be executed.
S905，云端通过共视判断，确定新获取的图像能够用于更新地图。S905: The cloud determines through common-view judgment that the newly acquired image can be used to update the map.
其中，这里新获取的图像可以指在统计周期内，云端获取的GPS区域的query图像。The newly acquired image here may refer to a query image of the GPS area acquired by the cloud within the statistical period.
为了便于理解，可以将步骤S905涉及的过程分为三个阶段：阶段1，云端构建场景图；阶段2，云端确定场景图连通；阶段3，统计场景图中正常定位图像的数量达到或超过一定数目。For ease of understanding, the process involved in step S905 can be divided into three stages: stage 1, the cloud builds a scene graph; stage 2, the cloud determines that the scene graph is connected; stage 3, the number of normally positioned images in the scene graph reaches or exceeds a certain number.
首先，针对阶段1，需要说明的是，同一个场景图中的图像是具有共视关系的。构建场景图的原理可以理解为，如果两张图像之间具有共视关系，有可以匹配的特征，那么以特征匹配数目作为边，可以将两个节点连接起来，这两个节点可以对应上述两张图像。类似地，可以把该GPS区域中具有共视关系的图像都作为节点，通过两两图像可以匹配的特征数目作为边，将两两图像连接起来，最终获取节点全部为具有连接关系的完整场景图。示例性的，该场景图对应的模型可以如图10所示。First, for stage 1, it should be noted that the images in the same scene graph have a co-viewing relationship. The principle of constructing a scene graph can be understood as follows: if there is a co-viewing relationship between two images and there are matching features, then the two nodes can be connected using the number of feature matches as an edge, and these two nodes can correspond to the above two images. Similarly, all images with a co-viewing relationship in the GPS area can be used as nodes, and the number of features that can be matched between the two images can be used as edges to connect the two images, and finally obtain a complete scene graph in which all nodes have a connection relationship. Exemplarily, the model corresponding to the scene graph can be shown in Figure 10.
示例性的，云端构建场景图的具体过程可以包括：(1)云端对该GPS区域在异常时间之前的正常定位图像，以及在异常时间之后的异常定位图像进行特征提取，其中，这里的异常时间也即上文所述的该GPS区域的异常定位图像的占比明显增大的时间，也即对应于该GPS区域中一场定位图象的占比等于或大于第二阈值的时间。这里的特征例如可以是任何一种特异性较强的点特征，如利用尺度不变特征变换(scale-invariant featuretransform，SIFT)算法获取的图像点特征。(2)SIFT算法的匹配是基于外观的匹配，没有考虑图像之间的几何关系，不能保证对应的特征能映射到同一三维场景点，因此，云端接下来可以利用RANSAC的方法计算(正常定位图像和异常定位图像)两两图像的基础矩阵和单应性矩阵，根据几何关系筛选出可靠度比较高的匹配对，剔除外点。(3)云端输出该GPS区域对应的完整场景图，该场景图的节点是图像，边是图像特征匹配的数目。Exemplarily, the specific process of constructing a scene graph in the cloud may include: (1) The cloud extracts features from the normal positioning image of the GPS area before the abnormal time and the abnormal positioning image after the abnormal time, wherein the abnormal time here is the time when the proportion of abnormal positioning images in the GPS area mentioned above increases significantly, that is, the time when the proportion of a field positioning image in the GPS area is equal to or greater than the second threshold. The features here can be, for example, any point feature with strong specificity, such as image point features obtained using the scale-invariant feature transform (SIFT) algorithm. (2) The matching of the SIFT algorithm is based on appearance matching, without considering the geometric relationship between images, and cannot guarantee that the corresponding features can be mapped to the same three-dimensional scene point. Therefore, the cloud can then use the RANSAC method to calculate the basic matrix and homography matrix of each image (normal positioning image and abnormal positioning image), and select matching pairs with relatively high reliability based on the geometric relationship, and eliminate outliers. (3) The cloud outputs a complete scene graph corresponding to the GPS area, where the nodes of the scene graph are images and the edges are the number of image feature matches.
针对阶段2，云端确定场景图连通的过程可以包括：(1)统计场景图包括的两种类型图像(也即正常定位图像和异常定位图像)的数量。(2)如果场景图包括的正常定位图像的数量和异常定位图像的数量分别达到预设数量，例如场景图包括的正常定位图像的数量等于或大于第三阈值，以及场景图包括的异常定位图像的数量等于或大于第四阈值，则意味着正常定位图像和异常定位图像之间具有一定数量的共性的特征，而不是全部为相同类型的图像连接，或者一种类型的图像连接数量远远大于两种不同类型图像的连接数量。For stage 2, the process of determining the connectivity of the scene graph in the cloud may include: (1) counting the number of two types of images (i.e., normal positioning images and abnormal positioning images) included in the scene graph. (2) If the number of normal positioning images and the number of abnormal positioning images included in the scene graph reach a preset number, for example, the number of normal positioning images included in the scene graph is equal to or greater than the third threshold, and the number of abnormal positioning images included in the scene graph is equal to or greater than the fourth threshold, it means that the normal positioning images and the abnormal positioning images have a certain number of common features, rather than all being connected by the same type of images, or the number of image connections of one type is much greater than the number of connections of two different types of images.
判断场景图的连通性，是为了保证真实环境变化之后，异常定位图像和原来正常定位图像有一定的关联性，从而保证这些异常定位图像就是在该GPS区域采集到的，环境变化的确是发生在该GPS区域的，而不是将其他环境的图像误用到分析当前GPS区域的地图变化。The purpose of judging the connectivity of the scene graph is to ensure that after the real environment changes, there is a certain correlation between the abnormal positioning images and the original normal positioning images, thereby ensuring that these abnormal positioning images are collected in the GPS area and that the environmental changes do occur in the GPS area, rather than misusing images of other environments to analyze map changes in the current GPS area.
针对阶段3，需要说明的是，云端之所以需要统计场景图中正常定位图像占比是否达到或超过一定比例(如第五阈值)，是因为从实际情况来说，即使某个GPS区域的真实环境发生变化，通常变化的部分会少于或者远少于未变化的部分，从这个角度来说，该阶段3的操作可以辅助证明数据是否可靠。此外，该操作更为关键的目的是，为了保证新的部分最终能够更好地替换离线三维模型中原有的对应部分，也即更好地融合到原本的点云地图中，而非生硬地加入。换句话说，本申请实施例中的场景图对应于一个GPS区域，而该GPS区域可能仅对应于离线三维模型的一小部分，然而点云地图更新的最终目的是将该GPS区域的变化的部分融合到已建图区域对应的完整离线三维模型中，因而如果场景图中正常定位图像比例太小，那么最终进行融合时，就会存在较大误差，无法获得较好的融合效果。Regarding stage 3, it should be noted that the reason why the cloud needs to count whether the proportion of normal positioning images in the scene graph reaches or exceeds a certain proportion (such as the fifth threshold) is because, in actual conditions, even if the real environment of a certain GPS area changes, the changed part will usually be less than or far less than the unchanged part. From this perspective, the operation of stage 3 can assist in proving whether the data is reliable. In addition, the more critical purpose of this operation is to ensure that the new part can eventually better replace the original corresponding part in the offline three-dimensional model, that is, better integrate into the original point cloud map, rather than rigidly join. In other words, the scene graph in the embodiment of the present application corresponds to a GPS area, and the GPS area may only correspond to a small part of the offline three-dimensional model. However, the ultimate purpose of updating the point cloud map is to integrate the changed part of the GPS area into the complete offline three-dimensional model corresponding to the mapped area. Therefore, if the proportion of normal positioning images in the scene graph is too small, there will be a large error when the fusion is finally performed, and a good fusion effect cannot be obtained.
其中，阶段2和阶段3中判断过程所依据的是上文所述的第二条件。也即，第二条件可以包括：场景图包括的两种类型图像(也即正常定位图像和异常定位图像)的数量是否分别达到预设数量，比如场景图包括的正常定位图像的数量等于或大于第三阈值，以及场景图包括的异常定位图像的数量等于或大于第四阈值；以及场景图中正常定位图像的占比是否达到或超过一定比例，比如场景图中正常定位图像的占比等于或大于第五阈值。Among them, the judgment process in stage 2 and stage 3 is based on the second condition described above. That is, the second condition may include: whether the number of two types of images (i.e., normal positioning images and abnormal positioning images) included in the scene graph reaches a preset number respectively, such as the number of normal positioning images included in the scene graph is equal to or greater than the third threshold, and the number of abnormal positioning images included in the scene graph is equal to or greater than the fourth threshold; and whether the proportion of normal positioning images in the scene graph reaches or exceeds a certain proportion, such as the proportion of normal positioning images in the scene graph is equal to or greater than the fifth threshold.
S906，云端对GPS区域对应的点云地图进行更新。S906, the cloud updates the point cloud map corresponding to the GPS area.
在一些实施例中，云端进行地图更新的过程可以包括：将GPS区域对应的多个不同视角下图像的同一位置作为特征点，对这些特征点进行三角化处理，将特征点信息转化为三维点云信息，之后，基于该三维点云信息对离线三维模型中对应的部分进行更新，即可获取更新后的点云地图。In some embodiments, the process of updating the map in the cloud may include: taking the same position of images under multiple different perspectives corresponding to the GPS area as feature points, triangulating these feature points, converting the feature point information into three-dimensional point cloud information, and then updating the corresponding part in the offline three-dimensional model based on the three-dimensional point cloud information to obtain an updated point cloud map.
根据本申请实施例提供的点云地图更新的方法，通过云端与终端设备的交互获取已建图区域的图像，并基于VPS定位技术判断对点云地图更新的必要性，能够实现无需过度耗费人力，不仅极大地节约了地图更新的成本，还更加智能化地提升了地图更新时机的准确性，保持了地图鲜度。此外，本方法在更新之前基于构建的场景图进行点云地图更新融合性的判断，能够保证更新部分更好地融合到原有的离线三维模型中，获取融合效果更好的点云地图。According to the method for updating the point cloud map provided in the embodiment of the present application, the image of the mapped area is obtained through the interaction between the cloud and the terminal device, and the necessity of updating the point cloud map is judged based on the VPS positioning technology, which can achieve the goal of not over-consuming manpower, greatly saving the cost of map updates, and more intelligently improving the accuracy of the timing of map updates, thereby maintaining the freshness of the map. In addition, before updating, the method judges the fusion of the point cloud map update based on the constructed scene graph, which can ensure that the updated part is better integrated into the original offline three-dimensional model, and obtain a point cloud map with better fusion effect.
示例性的，如图11所示，为本申请实施例提供的一种点云地图更新的方法的示意性流程图。该方法可以由云端作为执行主体，具体可以包括以下步骤：For example, as shown in FIG11 , a schematic flow chart of a method for updating a point cloud map provided in an embodiment of the present application is provided. The method can be executed by the cloud and can specifically include the following steps:
S1101，获取终端设备发送的图像信息，该图像信息包括第一区域的图像和第一区域的图像对应的位置信息。S1101, obtaining image information sent by a terminal device, where the image information includes an image of a first area and position information corresponding to the image of the first area.
其中，该第一区域可以对应于上文实施例中的GPS区域，如第一GPS区域。The first area may correspond to the GPS area in the above embodiment, such as the first GPS area.
本申请主要以GPS作为示例进行介绍，但在实际应用中，本方法中的位置信息也可以通过其他方式获得，比如通过北斗卫星定位系统等，本申请实施例对此不作限定。This application mainly introduces GPS as an example, but in actual applications, the location information in this method can also be obtained through other means, such as through the Beidou satellite positioning system, etc., and the embodiments of this application are not limited to this.
S1102，根据第一区域的图像对应的位置信息，通过VPS获取图像的定位结果，其中，定位结果准确的图像为正常定位图像，定位结果不准确的图像为异常定位图像。S1102, obtaining the positioning result of the image through the VPS according to the position information corresponding to the image of the first area, wherein the image with accurate positioning result is a normal positioning image, and the image with inaccurate positioning result is an abnormal positioning image.
S1103，根据异常定位图像的数量获取是否需要更新第一区域对应的点云地图。S1103: Determine whether it is necessary to update the point cloud map corresponding to the first area according to the number of abnormal positioning images.
S1104，当需要更新第一区域对应的点云地图时，获取第一区域对应的场景图，该场景图的节点为与其他图像具有特征匹配关系的正常定位图像和异常定位图像，该场景图的边为节点对应的图像之间匹配的特征数目。S1104, when it is necessary to update the point cloud map corresponding to the first area, obtain a scene graph corresponding to the first area, the nodes of the scene graph are normal positioning images and abnormal positioning images that have feature matching relationships with other images, and the edges of the scene graph are the number of features matched between the images corresponding to the nodes.
S1105，根据场景图中正常定位图像的数量和异常定位图像的数量，更新第一区域对应的点云地图。S1105: Update the point cloud map corresponding to the first area according to the number of normal positioning images and the number of abnormal positioning images in the scene graph.
在一些实施例中，所述方法还包括：存储离线三维模型，所述离线三维模型对应的区域为已建图区域；其中，所述离线三维模型包括第一部分和其他部分，所述第一部分为未更新的所述第一区域对应的点云地图，所述其他部分为所述已建图区域中剩余其他区域对应的点云地图。In some embodiments, the method further includes: storing an offline three-dimensional model, the area corresponding to the offline three-dimensional model being a mapped area; wherein the offline three-dimensional model includes a first part and other parts, the first part being a point cloud map corresponding to the first area that has not been updated, and the other parts being point cloud maps corresponding to the remaining other areas in the mapped area.
在一些实施例中，所述图像信息还包括所述其他区域的图像和所述其他区域的图像对应的位置信息，所述方法还包括：根据所述图像信息中的位置信息和所述第一区域的位置，进行图像分类，获取属于所述第一区域的图像。In some embodiments, the image information also includes images of the other areas and location information corresponding to the images of the other areas, and the method also includes: classifying the images according to the location information in the image information and the location of the first area to obtain images belonging to the first area.
在一些实施例中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：若所述异常定位图像的数量等于或大于第一阈值，则确定需要更新所述第一区域对应的点云地图。In some embodiments, obtaining whether it is necessary to update the point cloud map corresponding to the first area based on the number of abnormal positioning images specifically includes: if the number of the abnormal positioning images is equal to or greater than a first threshold, determining that the point cloud map corresponding to the first area needs to be updated.
在一些实施例中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：若所述异常定位图像的数量等于或大于第一阈值，且在某个异常时间之后，所述异常定位图像的占比等于或大于第二阈值，则确定需要更新所述第一区域对应的点云地图。In some embodiments, the step of obtaining whether it is necessary to update the point cloud map corresponding to the first area based on the number of abnormal positioning images specifically includes: if the number of the abnormal positioning images is equal to or greater than a first threshold, and after a certain abnormal time, the proportion of the abnormal positioning images is equal to or greater than a second threshold, then it is determined that the point cloud map corresponding to the first area needs to be updated.
在一些实施例中，所述根据异常定位图像的数量获取是否需要更新所述第一区域对应的点云地图，具体包括：若所述异常定位图像的数量等于或大于第一阈值；以及，在某个异常时间之后，所述异常定位图像的占比等于或大于第二阈值；以及，所述正常定位图像和所述异常定位图像在语义分割后的语义类别差距大于预设差距，则确定需要更新所述第一区域对应的点云地图。In some embodiments, the step of obtaining whether the point cloud map corresponding to the first area needs to be updated based on the number of abnormal positioning images specifically includes: if the number of the abnormal positioning images is equal to or greater than a first threshold; and, after a certain abnormal time, the proportion of the abnormal positioning images is equal to or greater than a second threshold; and, if the semantic category gap between the normal positioning image and the abnormal positioning image after semantic segmentation is greater than a preset gap, it is determined that the point cloud map corresponding to the first area needs to be updated.
在一些实施例中，所述当需要更新所述第一区域对应的点云地图时，获取所述第一区域对应的场景图，具体包括：对所述第一区域在异常时间之前的所述正常定位图像和所述异常时间之后的异常定位图像进行共视判断；将共视判断后与其他图像具有共同特征的图像作为所述节点，将两两图像之间匹配的特征数目作为所述两两图像对应节点之间的边，获取所述场景图。In some embodiments, when the point cloud map corresponding to the first area needs to be updated, a scene graph corresponding to the first area is obtained, specifically including: performing a common-view judgment on the normal positioning image of the first area before the abnormal time and the abnormal positioning image after the abnormal time; using the image that has common features with other images after the common-view judgment as the node, and using the number of matching features between the two images as the edge between the nodes corresponding to the two images, to obtain the scene graph.
在一些实施例中，所述根据所述场景图中所述正常定位图像的数量和所述异常定位图像的数量，对所述第一区域对应的点云地图进行更新，具体包括：当所述场景图包括的所述正常定位图像的数量等于或大于第三阈值，以及所述场景图包括的异常定位图像的数量等于或大于第四阈值时，统计所述场景图包括的所述正常定位图像的占比是否等于或大于第五阈值；若所述场景图包括的所述正常定位图像的占比是否等于或大于第五阈值，则对所述第一区域对应的点云地图进行更新。In some embodiments, the point cloud map corresponding to the first area is updated according to the number of normal positioning images and the number of abnormal positioning images in the scene graph, specifically including: when the number of normal positioning images included in the scene graph is equal to or greater than a third threshold, and the number of abnormal positioning images included in the scene graph is equal to or greater than a fourth threshold, counting whether the proportion of the normal positioning images included in the scene graph is equal to or greater than a fifth threshold; if the proportion of the normal positioning images included in the scene graph is equal to or greater than the fifth threshold, updating the point cloud map corresponding to the first area.
在一些实施例中，所述对所述第一区域对应的点云地图进行更新，具体包括：对所述第一区域对应的所述正常定位图像和所述异常定位图像均包括的同一位置作为特征点；对所述特征点进行三角化处理，获取与所述特征点对应的三维点云信息；在所述离线三维模型中，基于所述三维点云信息对所述第一区域对应的点云地图进行更新。In some embodiments, the updating of the point cloud map corresponding to the first area specifically includes: taking the same position included in the normal positioning image and the abnormal positioning image corresponding to the first area as a feature point; triangulating the feature point to obtain three-dimensional point cloud information corresponding to the feature point; and in the offline three-dimensional model, updating the point cloud map corresponding to the first area based on the three-dimensional point cloud information.
在一些实施例中，所述图像信息还包括时间信息；所述根据所述图像信息中的位置信息和所述第一区域的位置，进行图像分类，获取属于所述第一区域的图像，还包括：所述根据所述图像信息中时间信息，进行图像分类，获取属于所述第一区域的且在同一时间段内的图像。In some embodiments, the image information also includes time information; the image classification is performed based on the position information in the image information and the position of the first area to obtain images belonging to the first area, and also includes: the image classification is performed based on the time information in the image information to obtain images belonging to the first area and within the same time period.
根据本申请实施例提供的点云地图更新的方法，通过云端与终端设备的交互获取已建图区域的图像，并基于VPS定位技术判断对点云地图更新的必要性，能够实现无需过度耗费人力，不仅极大地节约了地图更新的成本，还更加智能化地提升了地图更新时机的准确性，保持了地图鲜度。此外，本方法在更新之前基于构建的场景图进行点云地图更新融合性的判断，能够保证更新部分更好地融合到原有的离线三维模型中，获取融合效果更好的点云地图。According to the method for updating the point cloud map provided in the embodiment of the present application, the image of the mapped area is obtained through the interaction between the cloud and the terminal device, and the necessity of updating the point cloud map is judged based on the VPS positioning technology, which can achieve the goal of not over-consuming manpower, greatly saving the cost of map updates, and more intelligently improving the accuracy of the timing of map updates, thereby maintaining the freshness of the map. In addition, before updating, the method judges the fusion of the point cloud map update based on the constructed scene graph, which can ensure that the updated part is better integrated into the original offline three-dimensional model, and obtain a point cloud map with better fusion effect.
基于同样的技术构思，本申请实施例还提供了一种云端设备，包括一个或多个处理器；一个或多个存储器；所述一个或多个存储器存储有一个或多个计算机程序，所述一个或多个计算机程序包括指令，当所述指令被所述一个或多个处理器执行时，使得计算机或处理器执行上述任一个方法中的一个或多个步骤。Based on the same technical concept, an embodiment of the present application also provides a cloud device, including one or more processors; one or more memories; the one or more memories store one or more computer programs, and the one or more computer programs include instructions. When the instructions are executed by the one or more processors, the computer or processor executes one or more steps in any of the above methods.
基于同样的技术构思，本申请实施例还提供了一种终端设备，包括一个或多个处理器；一个或多个存储器；所述一个或多个存储器存储有一个或多个计算机程序，所述一个或多个计算机程序包括指令，当所述指令被所述一个或多个处理器执行时，使得计算机或处理器执行上述任一个方法中的一个或多个步骤。Based on the same technical concept, an embodiment of the present application also provides a terminal device, including one or more processors; one or more memories; the one or more memories store one or more computer programs, and the one or more computer programs include instructions. When the instructions are executed by the one or more processors, the computer or processor executes one or more steps in any of the above methods.
基于同样的技术构思，本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有计算机可执行程序指令，所述计算机可执行程序指令在被计算机上运行时，使得计算机或处理器执行上述任一个方法中的一个或多个步骤。Based on the same technical concept, an embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable program instructions. When the computer-executable program instructions are executed on a computer, the computer or processor executes one or more steps in any of the above methods.
基于同样的技术构思，本申请实施例还提供了一种包含指令的计算机程序产品，所述计算机程序产品包括计算机程序代码，当所述计算机程序代码在计算机上运行时，使得计算机或处理器执行上述任一个方法中的一个或多个步骤。Based on the same technical concept, an embodiment of the present application also provides a computer program product comprising instructions, wherein the computer program product includes computer program code. When the computer program code runs on a computer, the computer or processor executes one or more steps in any of the above methods.
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其它可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者通过所述计算机可读存储介质进行传输。所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如，固态硬盘(solid state disk，SSD))等。In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，该流程可以由计算机程序来指令相关的硬件完成，该程序可存储于计算机可读取存储介质中，该程序在执行时，可包括如上述各方法实施例的流程。而前述的存储介质包括：ROM或随机存储记忆体RAM、磁碟或者光盘等各种可存储程序代码的介质。A person skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by a computer program to instruct the relevant hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The aforementioned storage medium includes: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.
以上所述，仅为本申请实施例的具体实施方式，但本申请实施例的保护范围并不局限于此，任何在本申请实施例揭露的技术范围内的变化或替换，都应涵盖在本申请实施例的保护范围之内。因此，本申请实施例的保护范围应以所述权利要求的保护范围为准。The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the embodiment of the present application should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477578.5ACN117128985B (en) | 2023-04-27 | 2023-04-27 | Point cloud map updating method and equipment |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477578.5ACN117128985B (en) | 2023-04-27 | 2023-04-27 | Point cloud map updating method and equipment |
| Publication Number | Publication Date |
|---|---|
| CN117128985A CN117128985A (en) | 2023-11-28 |
| CN117128985Btrue CN117128985B (en) | 2024-05-31 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310477578.5AActiveCN117128985B (en) | 2023-04-27 | 2023-04-27 | Point cloud map updating method and equipment |
| Country | Link |
|---|---|
| CN (1) | CN117128985B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117710595A (en)* | 2023-12-13 | 2024-03-15 | 江苏徐工工程机械研究院有限公司 | Point cloud map updating method and device, engineering vehicle and storage medium |
| US20250272915A1 (en)* | 2024-02-22 | 2025-08-28 | Google Llc | Scanning framework for mapping a space |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360245A (en)* | 2018-10-26 | 2019-02-19 | 魔视智能科技(上海)有限公司 | Extrinsic parameter calibration method for multi-camera system of unmanned vehicle |
| CN112927362A (en)* | 2021-04-07 | 2021-06-08 | Oppo广东移动通信有限公司 | Map reconstruction method and device, computer readable medium and electronic device |
| CN113052152A (en)* | 2021-06-02 | 2021-06-29 | 中国人民解放军国防科技大学 | Indoor semantic map construction method, device and equipment based on vision |
| CN113537208A (en)* | 2021-05-18 | 2021-10-22 | 杭州电子科技大学 | Visual positioning method and system based on semantic ORB-SLAM technology |
| CN113744308A (en)* | 2021-08-06 | 2021-12-03 | 高德软件有限公司 | Pose optimization method, pose optimization device, electronic device, pose optimization medium, and program product |
| CN113776544A (en)* | 2020-06-10 | 2021-12-10 | 杭州海康威视数字技术股份有限公司 | Point cloud map updating method and device, electronic equipment and positioning system |
| WO2022063056A1 (en)* | 2020-09-22 | 2022-03-31 | 重庆兰德适普信息科技有限公司 | Method for constructing high-precision point cloud map |
| CN114332394A (en)* | 2021-12-29 | 2022-04-12 | 北京航空航天大学 | Semantic information assistance-based dynamic scene three-dimensional reconstruction method |
| CN114969221A (en)* | 2021-02-20 | 2022-08-30 | 华为技术有限公司 | A method for updating a map and related equipment |
| WO2023050647A1 (en)* | 2021-09-28 | 2023-04-06 | 上海仙途智能科技有限公司 | Map updating method and apparatus, computer device, and medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109360245A (en)* | 2018-10-26 | 2019-02-19 | 魔视智能科技(上海)有限公司 | Extrinsic parameter calibration method for multi-camera system of unmanned vehicle |
| CN113776544A (en)* | 2020-06-10 | 2021-12-10 | 杭州海康威视数字技术股份有限公司 | Point cloud map updating method and device, electronic equipment and positioning system |
| WO2022063056A1 (en)* | 2020-09-22 | 2022-03-31 | 重庆兰德适普信息科技有限公司 | Method for constructing high-precision point cloud map |
| CN114969221A (en)* | 2021-02-20 | 2022-08-30 | 华为技术有限公司 | A method for updating a map and related equipment |
| CN112927362A (en)* | 2021-04-07 | 2021-06-08 | Oppo广东移动通信有限公司 | Map reconstruction method and device, computer readable medium and electronic device |
| CN113537208A (en)* | 2021-05-18 | 2021-10-22 | 杭州电子科技大学 | Visual positioning method and system based on semantic ORB-SLAM technology |
| CN113052152A (en)* | 2021-06-02 | 2021-06-29 | 中国人民解放军国防科技大学 | Indoor semantic map construction method, device and equipment based on vision |
| CN113744308A (en)* | 2021-08-06 | 2021-12-03 | 高德软件有限公司 | Pose optimization method, pose optimization device, electronic device, pose optimization medium, and program product |
| WO2023050647A1 (en)* | 2021-09-28 | 2023-04-06 | 上海仙途智能科技有限公司 | Map updating method and apparatus, computer device, and medium |
| CN114332394A (en)* | 2021-12-29 | 2022-04-12 | 北京航空航天大学 | Semantic information assistance-based dynamic scene three-dimensional reconstruction method |
| Publication number | Publication date |
|---|---|
| CN117128985A (en) | 2023-11-28 |
| Publication | Publication Date | Title |
|---|---|---|
| CN109919251B (en) | Image-based target detection method, model training method and device | |
| CN112784174B (en) | Method, device and system for determining pose | |
| CN117128985B (en) | Point cloud map updating method and equipment | |
| CN107332981B (en) | Image processing method and related products | |
| CN111429517A (en) | Relocation method, relocation device, storage medium and electronic device | |
| CN110495819B (en) | Robot control method, robot, terminal, server and control system | |
| US20220124597A1 (en) | Method for Identifying Specific Position on Specific Route and Electronic Device | |
| CN110381195A (en) | A kind of throwing screen display methods and electronic equipment | |
| CN112270754A (en) | Local grid map construction method and device, readable medium and electronic device | |
| CN112785700B (en) | Virtual object display method, global map updating method and equipment | |
| CN106389078A (en) | Intelligent blind guiding glass system and blind guiding method thereof | |
| CN103533248A (en) | Image processing method, terminal and system | |
| CN108989665A (en) | Image processing method, device, mobile terminal and computer-readable medium | |
| CN115937722A (en) | A device positioning method, device and system | |
| CN114466449B (en) | Position feature obtaining method and terminal equipment | |
| CN114861032A (en) | Searching method and electronic equipment | |
| CN113128265B (en) | Character recognition method and device | |
| CN112700525B (en) | Image processing method and electronic device | |
| CN116029952B (en) | Point cloud evaluation methods and related equipment | |
| WO2023216957A1 (en) | Target positioning method and system, and electronic device | |
| CN114283195B (en) | Method, electronic device and readable storage medium for generating dynamic images | |
| CN115333941A (en) | Method and related device for obtaining running status of application | |
| CN116709180A (en) | Geographic fence generation method and server | |
| CN116561085A (en) | Image sharing methods and electronic devices | |
| WO2023024036A1 (en) | Method and apparatus for reconstructing three-dimensional model of person |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | Address after:Unit 3401, unit a, building 6, Shenye Zhongcheng, No. 8089, Hongli West Road, Donghai community, Xiangmihu street, Futian District, Shenzhen, Guangdong 518040 Patentee after:Honor Terminal Co.,Ltd. Country or region after:China Address before:3401, unit a, building 6, Shenye Zhongcheng, No. 8089, Hongli West Road, Donghai community, Xiangmihu street, Futian District, Shenzhen, Guangdong Patentee before:Honor Device Co.,Ltd. Country or region before:China | |
| CP03 | Change of name, title or address |