技术领域technical field
本发明涉及神经网络技术领域，尤其涉及一种基于历史视频的导航方法、装置、计算机设备及存储介质。The invention relates to the technical field of neural networks, in particular to a historical video-based navigation method, device, computer equipment and storage medium.
背景技术Background technique
目前，因城市交通路线越来越错综复杂，导航软件得到了越来越广泛的应用。现有的导航软件在进行导航时，一般是简单的二维平面地图结合道路实景图来对用户进行路线导航，无法进行恶劣天气情况下的辅助导航。At present, due to the increasingly complicated urban traffic routes, navigation software has been more and more widely used. Existing navigation software generally uses a simple two-dimensional planar map combined with a real road map to guide the user when performing navigation, and cannot provide auxiliary navigation under severe weather conditions.
发明内容Contents of the invention
本发明实施例提供了一种基于历史视频的导航方法、装置、计算机设备及存储介质，旨在解决现有技术中导航软件在进行导航时，一般是简单的二维平面地图结合道路实景图来对用户进行路线导航，无法进行恶劣天气情况下的辅助导航的问题。Embodiments of the present invention provide a historical video-based navigation method, device, computer equipment, and storage medium, aiming to solve the problem of navigation software in the prior art, which is generally a simple two-dimensional planar map combined with a real road map. Route navigation for users, the problem that auxiliary navigation cannot be performed under severe weather conditions.
第一方面，本发明实施例提供了一种基于历史视频的导航方法，其包括：In the first aspect, the embodiment of the present invention provides a navigation method based on historical video, which includes:
若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据；If the time interval between the current moment and the previous video acquisition moment is equal to the preset acquisition cycle, obtain the road traffic video data of the road corresponding to the preset geographic location coordinates at the current moment;
将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片；Carrying out video decomposition of the road traffic condition video data to obtain multi-frame road traffic condition pictures corresponding to the road traffic condition video data;
在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片；Obtaining a frame of pictures at random from the multi-frame road traffic picture corresponding to the road traffic video data, as the actual road traffic picture corresponding to the road traffic video data;
将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果；Using the real picture of the actual road condition as the input of the pre-trained road weather condition recognition model to obtain the road weather condition recognition result of the real picture of the actual road condition;
若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标；以及If the road weather condition identification result is greater than the preset severe weather warning value, obtain the corresponding actual road condition real picture and the geographic location coordinates corresponding to the actual road condition real picture; and
将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。The real picture of the actual road condition corresponding to the geographic location coordinates is replaced with the real picture of the road condition in non-bad weather conditions.
第二方面，本发明实施例提供了一种基于历史视频的导航装置，其包括：In a second aspect, an embodiment of the present invention provides a navigation device based on historical video, which includes:
当前视频采集单元，用于若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据；The current video acquisition unit is used to obtain the road traffic video data of the road corresponding to the preset geographical location coordinates at the current moment if the time interval between the current moment and the previous video acquisition moment is equal to the preset acquisition cycle;
当前视频分解单元，用于将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片；The current video decomposition unit is configured to perform video decomposition on the road traffic condition video data to obtain multi-frame road traffic condition pictures corresponding to the road traffic condition video data;
随机选取单元，用于在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片；A random selection unit, configured to randomly obtain a frame of pictures from the multi-frame road condition pictures corresponding to the road condition video data, as an actual road condition scene picture corresponding to the road condition video data;
天气状况识别单元，用于将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果；The weather condition recognition unit is used to use the actual road condition real picture as the input of the pre-trained road weather condition recognition model to obtain the road weather condition recognition result corresponding to the actual road condition real picture;
结果判断单元，用于若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标；以及A result judging unit, configured to obtain a corresponding actual road condition real picture and the geographic location coordinates corresponding to the actual road condition real picture if the road weather condition identification result is greater than the preset severe weather warning value; and
图片替换单元，用于将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。The picture replacing unit is used to replace the real picture of the actual road condition corresponding to the geographical location coordinates with the real picture of the road condition under non-bad weather conditions.
第三方面，本发明实施例又提供了一种计算机设备，其包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述第一方面所述的基于历史视频的导航方法。In a third aspect, the embodiment of the present invention provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor executes the computer program. The program implements the historical video-based navigation method described in the first aspect above.
第四方面，本发明实施例还提供了一种计算机可读存储介质，其中所述计算机可读存储介质存储有计算机程序，所述计算机程序当被处理器执行时使所述处理器执行上述第一方面所述的基于历史视频的导航方法。In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the above-mentioned first step. In one aspect, the historical video-based navigation method.
本发明实施例提供了一种基于历史视频的导航方法、装置、计算机设备及存储介质。该方法包括若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据；将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片；在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片；将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果；若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标；以及将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。该方法采用神经网络技术，实现了按周期获取道路路况图片，并且及时更新到电子地图上，基于实时路况数据，在复杂路面情况下时可有效导航。The embodiment of the present invention provides a historical video-based navigation method, device, computer equipment and storage medium. The method comprises that if the time interval between the current moment and the previous video collection moment is equal to the preset collection period, acquiring the road traffic video data of the road corresponding to the preset geographical position coordinates at the current moment; and performing the road traffic video data on the road The video is decomposed to obtain a multi-frame road traffic picture corresponding to the road traffic video data; a frame of picture is randomly obtained in the multi-frame road traffic picture corresponding to the road traffic video data, as the image corresponding to the road traffic video data Corresponding actual road condition real picture; with described actual road condition real picture as the input of the road weather condition recognition model of pre-training, obtain the road weather condition recognition result with described actual road condition real picture; If described road weather condition The identification result is greater than the preset severe weather warning value, and the corresponding actual road condition real picture is obtained, and the corresponding geographical position coordinates of the actual road condition real picture; and the actual road condition real picture corresponding to the geographical position coordinate is replaced by Real pictures of road conditions in non-bad weather conditions. The method adopts the neural network technology to obtain road condition pictures periodically and update them to the electronic map in time. Based on the real-time traffic condition data, it can effectively navigate under complex road conditions.
附图说明Description of drawings
为了更清楚地说明本发明实施例技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. Ordinary technicians can also obtain other drawings based on these drawings on the premise of not paying creative work.
图1为本发明实施例提供的基于历史视频的导航方法的应用场景示意图；FIG. 1 is a schematic diagram of an application scenario of a navigation method based on historical video provided by an embodiment of the present invention;
图2为本发明实施例提供的基于历史视频的导航方法的流程示意图；FIG. 2 is a schematic flow chart of a navigation method based on historical video provided by an embodiment of the present invention;
图3为本发明实施例提供的基于历史视频的导航方法的子流程示意图；FIG. 3 is a schematic subflow diagram of a navigation method based on historical video provided by an embodiment of the present invention;
图4为本发明实施例提供的基于历史视频的导航装置的示意性框图；FIG. 4 is a schematic block diagram of a navigation device based on historical video provided by an embodiment of the present invention;
图5为本发明实施例提供的基于历史视频的导航装置的子单元示意性框图；FIG. 5 is a schematic block diagram of subunits of a historical video-based navigation device provided by an embodiment of the present invention;
图6为本发明实施例提供的计算机设备的示意性框图。Fig. 6 is a schematic block diagram of a computer device provided by an embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
应当理解，当在本说明书和所附权利要求书中使用时，术语“包括”和“包含”指示所描述特征、整体、步骤、操作、元素和/或组件的存在，但并不排除一个或多个其它特征、整体、步骤、操作、元素、组件和/或其集合的存在或添加。It should be understood that when used in this specification and the appended claims, the terms "comprising" and "comprises" indicate the presence of described features, integers, steps, operations, elements and/or components, but do not exclude one or Presence or addition of multiple other features, integers, steps, operations, elements, components and/or collections thereof.
还应当理解，在此本发明说明书中所使用的术语仅仅是出于描述特定实施例的目的而并不意在限制本发明。如在本发明说明书和所附权利要求书中所使用的那样，除非上下文清楚地指明其它情况，否则单数形式的“一”、“一个”及“该”意在包括复数形式。It should also be understood that the terminology used in the description of the present invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the present invention. As used in this specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural referents unless the context clearly dictates otherwise.
还应当进一步理解，在本发明说明书和所附权利要求书中使用的术语“和/或”是指相关联列出的项中的一个或多个的任何组合以及所有可能组合，并且包括这些组合。It should also be further understood that the term "and/or" used in the description of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .
请参阅图1和图2，图1为本发明实施例提供的基于历史视频的导航方法的应用场景示意图，图2为本发明实施例提供的基于历史视频的导航方法的流程示意图，该基于历史视频的导航方法应用于服务器中，该方法通过安装于服务器中的应用软件进行执行。Please refer to Fig. 1 and Fig. 2, Fig. 1 is a schematic diagram of the application scene of the historical video-based navigation method provided by the embodiment of the present invention, Fig. 2 is a schematic flow chart of the historical video-based navigation method provided by the embodiment of the present invention, the historical video-based The video navigation method is applied in the server, and the method is executed by application software installed in the server.
如图2所示，该方法包括步骤S110～S160。As shown in FIG. 2, the method includes steps S110-S160.
S110、若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据。S110. If the time interval between the current moment and the previous video collection moment is equal to the preset collection period, acquire road traffic video data of the road corresponding to the preset geographic location coordinates at the current moment.
在本实施例中，对设置在道路路口的前端采集装置(如高清摄像头)设置一个采集周期，是为了周期性的对各被监控的道路进行路况视频记录，每间隔采集周期(例如设置采集周期为30分钟)所采集的道路路况视频数据的时长也是相同的(例如5秒钟)，并可以对所采集的道路路况视频数据进行处理后，得到最新路况的导航图片。In this embodiment, a collection cycle is set for the front-end collection device (such as a high-definition camera) arranged at a road intersection, in order to periodically record the road condition video of each monitored road, and every interval collection cycle (such as setting the collection cycle The duration of the road traffic video data collected for 30 minutes) is also the same (for example, 5 seconds), and after the collected road traffic video data can be processed, a navigation picture of the latest road conditions can be obtained.
S120、将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片。S120. Perform video decomposition on the road traffic condition video data to obtain multiple frames of road traffic condition pictures corresponding to the road traffic condition video data.
在本实施例中，由视频的形成过程可知，视频是由多帧图片组成，例如每一秒的视频能转化成24-30张图片。通过视频分解工具(如Adobe Premiere，在Adobe Premiere中剪辑出需要输出序列位图的片段，从文件菜单里输出成序列位图即可)即可将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片。In this embodiment, it can be seen from the video forming process that the video is composed of multiple frames of pictures, for example, every second of video can be converted into 24-30 pictures. Through the video decomposition tool (such as Adobe Premiere, in Adobe Premiere, edit the segment that needs to output the sequence bitmap, and output it as a sequence bitmap from the file menu) the video decomposition of the road traffic condition video data can be carried out to obtain the same Multiple frames of road traffic picture corresponding to the road traffic video data.
S130、在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片。S130. Randomly acquire a frame of pictures from the multiple frames of road condition pictures corresponding to the road condition video data, as an actual road condition scene picture corresponding to the road condition video data.
在本实施例中，在当前时刻所采集预设时长(5秒钟)的道路路况视频数据，进行视频分解后，得到了120-150张道路路况图片，此时可以在其中随机选择一张作为当前时刻的道路路况图片，这样针对每一路段每间隔采集周期采集到道路路况视频数据仅保留一张图片作为道路路况实景图片。In this embodiment, the road traffic condition video data of the preset duration (5 seconds) collected at the current moment is decomposed into 120-150 road traffic condition pictures, and one of them can be randomly selected as The road condition picture at the current moment, so that only one picture is reserved for the road condition video data collected at each interval collection period for each road section as the real picture of the road condition.
在一实施例中，步骤S130之后还包括：In one embodiment, after step S130, it also includes:
将当前时刻的实际道路路况实景图片及对应的采集时间进行存储。Store the real picture of the actual road condition at the current moment and the corresponding collection time.
在具体实施时，针对每一个被监控的道路，在每间隔所述采集周期所采集的道路路况视频数据经视频分解和随机挑选一帧图片后仅保留一张道路路况实景图片，则每一被监控的道路在多个道路路况视频采集周期所积累下来有多张道路路况实景图片。例如道路路况视频采集周期设置为30分钟，则针对每一被监控的道路一天有48张道路路况实景图片进行保存。一旦该道路的状态发生改变，均可及时的上传提供导航服务的服务器，由服务器及时的更新各被监控的道路的道路路况实景图片。During specific implementation, for each monitored road, only one road traffic real scene picture is reserved after video decomposition and random selection of a frame of the road traffic video data collected at each interval of the collection period, then each monitored road The monitored road has multiple real-life pictures of road conditions accumulated in multiple road condition video collection cycles. For example, if the collection cycle of the road condition video is set to 30 minutes, then 48 real pictures of the road condition will be saved for each monitored road a day. Once the state of the road changes, it can be uploaded in time to the server providing the navigation service, and the real picture of the road condition of each monitored road can be updated in time by the server.
通过这种周期性更新道路的路况实景图片，能较为及时的提醒使用导航服务的用户每一道路的实际道路情况。例如，某一时刻当前道路进入临时维护状态，此时刻对应获取的该道路的道路路况实景图片中能及时体现该道路进入了维修状态，可能导致拥堵，提示司机及时规划其他路线绕行。By periodically updating the road condition real picture of the road, the user using the navigation service can be reminded of the actual road condition of each road in a timely manner. For example, if the current road enters a temporary maintenance state at a certain moment, the road condition real picture corresponding to the acquired road at this moment can reflect that the road has entered the maintenance state in time, which may lead to congestion, prompting the driver to plan other detours in time.
S140、将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果。S140. Using the real picture of the actual road condition as an input of a pre-trained road weather condition recognition model to obtain a road weather condition recognition result corresponding to the real picture of the actual road condition.
在一实施例中，步骤S140之前还包括：In one embodiment, before step S140, it also includes:
获取多个训练数据，以对待训练的道路天气状况识别模型进行训练，得到用于识别道路天气状况识别结果的道路天气状况识别模型；其中，多个训练数据中每一训练数据以道路路况图片对应的图片特征向量作为待训练的道路天气状况识别模型的输入，并以道路天气状况识别结果作为待训练的道路天气状况识别模型的输出；其中，多个训练数据中每一训练数据对应的道路路况图片的采集时刻与当前时刻之差小于或等于预设的时间间隔阈值；每一训练数据对应的道路天气状况识别结果为预先标注的天气状况识别结果。Obtain a plurality of training data to train the road weather condition identification model to be trained to obtain a road weather condition identification model for identifying the result of road weather condition identification; wherein, each training data in the plurality of training data corresponds to a road condition picture The image feature vector of is used as the input of the road weather condition recognition model to be trained, and the road weather condition recognition result is used as the output of the road weather condition recognition model to be trained; wherein, the corresponding road condition of each training data in a plurality of training data The difference between the image collection time and the current time is less than or equal to the preset time interval threshold; the road weather condition recognition result corresponding to each training data is the pre-marked weather condition recognition result.
在本实施例中，预先训练一个道路天气状况识别模型时，可以选定当前道路多张天气状况下所拍摄的道路路况实景图片，如在雨天、下雪天、雾霾天、晴天、阴天5种天气状况下的道路路况实景图片，然后设置晴天对应的道路天气状况识别结果为1，阴天对应的道路天气状况识别结果为2，雨天对应的道路天气状况识别结果为3，下雪天对应的道路天气状况识别结果为4，雾霾天对应的道路天气状况识别结果为5。通过大量的各种天气状况下的道路路况实景图片对应的图片特征向量作为输入，且将对应设置的道路天气状况识别结果作为输出，对深度神经网络进训练得到道路天气状况识别模型。In this embodiment, when pre-training a road weather condition recognition model, the real picture of the road condition taken under the current road condition can be selected, such as rainy day, snowy day, haze day, sunny day, cloudy day Real-scene pictures of road conditions under 5 kinds of weather conditions, and then set the road weather condition recognition result corresponding to sunny days as 1, the road weather condition recognition result corresponding to cloudy days as 2, the road weather condition recognition result corresponding to rainy days as 3, and the road weather condition recognition result corresponding to snowy days as 3. The corresponding road weather condition identification result is 4, and the road weather condition identification result corresponding to the haze day is 5. Through a large number of picture feature vectors corresponding to real road conditions under various weather conditions as input, and the corresponding set road weather condition recognition results as output, the deep neural network is trained to obtain the road weather condition recognition model.
所采用的深度神经网络即DNN(Deep Neural Networks，简称DNN)，DNN可以理解为有很多隐藏层的神经网络。DNN内部的神经网络层可以分为三类，输入层，隐藏层和输出层,一般来说第一层是输入层，最后一层是输出层，而中间的层数都是隐藏层。层与层之间是全连接的，也就是说，第i层的任意一个神经元一定与第i+1层的任意一个神经元相连。The deep neural network used is DNN (Deep Neural Networks, DNN for short), and DNN can be understood as a neural network with many hidden layers. The neural network layers inside the DNN can be divided into three categories, the input layer, the hidden layer and the output layer. Generally speaking, the first layer is the input layer, the last layer is the output layer, and the middle layers are all hidden layers. The layers are fully connected, that is, any neuron in the i-th layer must be connected to any neuron in the i+1-th layer.
在一实施例中，如图3所示，步骤S140中将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，包括：In one embodiment, as shown in FIG. 3 , in step S140, the real picture of the actual road condition is used as the input of the pre-trained road weather condition recognition model, including:
S141、获取与所述实际道路路况实景图片对应的像素矩阵；S141. Obtain a pixel matrix corresponding to the actual picture of the actual road condition;
S142、将所述像素矩阵作为卷积神经网络模型中输入层的输入，对应得到多个特征图；S142. Using the pixel matrix as the input of the input layer in the convolutional neural network model, correspondingly obtain a plurality of feature maps;
S143、将多个特征图输入至卷积神经网络模型的池化层，得到每一特征图对应的最大值所对应一维行向量；S143. Input multiple feature maps to the pooling layer of the convolutional neural network model, and obtain a one-dimensional row vector corresponding to the maximum value corresponding to each feature map;
S144、将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量；S144. Input the one-dimensional row vector corresponding to the maximum value corresponding to each feature map to the fully connected layer of the convolutional neural network model, and obtain the picture feature vector corresponding to the real picture of the actual road condition;
S145、将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。S145. Using the picture feature vector as an input of a pre-trained road weather condition recognition model.
在本实施例中，每一帧道路路况实景图片均可以转化为像素矩阵(如256*256的像素矩阵)，将像素矩阵作为卷积神经网络模型中输入层的输入时，输入层通过卷积操作得到若干个Feature Map(Feature Map可以理解为特征图)，在设置好卷积窗口的大小后，通过卷积窗口对像素矩阵进行卷积将得到若干个列数为1的Feature Map。In this embodiment, each frame of road conditions real picture can be converted into a pixel matrix (such as a 256*256 pixel matrix), when the pixel matrix is used as the input of the input layer in the convolutional neural network model, the input layer passes through The operation obtains several Feature Maps (Feature Maps can be understood as feature maps). After setting the size of the convolution window, the pixel matrix is convoluted through the convolution window to obtain several Feature Maps with a column number of 1.
在池化层中，可采用从之前一维的Feature Map中提出最大的值。通过这种Pooling方式(即池化的方式)可以解决可变图片像素大小的输入问题(因为不管FeatureMap中有多少个值，只需要提取其中的最大值)，最终池化层的输出为各个Feature Map的最大值，即一个一维的向量。In the pooling layer, the largest value can be proposed from the previous one-dimensional Feature Map. Through this Pooling method (that is, pooling method), the input problem of variable image pixel size can be solved (because no matter how many values there are in FeatureMap, only the maximum value needs to be extracted), and the output of the final pooling layer is each Feature The maximum value of Map, that is, a one-dimensional vector.
最后将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量，即可将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。Finally, the one-dimensional row vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer of the convolutional neural network model, and the picture feature vector corresponding to the real picture of the actual road condition is obtained, and the picture feature can be vector as input to a pre-trained road weather condition recognition model.
S150、若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标。S150. If the identification result of the road weather condition is greater than the preset severe weather warning value, obtain the corresponding real picture of the actual road condition and the geographic location coordinates corresponding to the real picture of the actual road condition.
在本实施例中，当通过道路天气状况识别模型对所述实际道路路况实景图片进行识别时，若当前时刻所采集的道路路况实景图片对应的道路天气状况识别结果为3，超出了预设的恶劣天气预警值2，则表示当前道路此时刻的道路路况实景图片不适宜作为道路的实景导航图片。此时需定位存在道路天气状况识别结果大于预设的恶劣天气预警值所对应道路的地理位置坐标以判断是哪条道路的实景导航图片需要替换为非恶劣天气状况下道路路况实景图片(如晴天或阴天下的道路路况实景图片)，一般选择与当前时间间隔最近的非恶劣天气状况下道路路况实景图片进行替换。In this embodiment, when the road weather condition identification model is used to identify the actual road condition real picture, if the road weather condition identification result corresponding to the road condition real picture collected at the current moment is 3, exceeding the preset A severe weather warning value of 2 indicates that the real picture of the current road condition at this moment is not suitable as a real picture of the road for navigation. At this time, it is necessary to locate the geographic location coordinates of the road corresponding to the road weather condition identification result greater than the preset severe weather warning value to determine which road the real-scene navigation picture needs to be replaced with a real-scene picture of the road condition in non-severe weather conditions (such as a sunny day) or a real picture of the road condition under cloudy sky), generally select the real picture of the road condition under the non-bad weather condition closest to the current time interval to replace.
S160、将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。S160. Replace the real picture of the actual road condition corresponding to the geographic location coordinates with the real picture of the road condition in non-bad weather conditions.
在一实施例中，步骤S160之前还包括：In one embodiment, before step S160, it also includes:
获取与所述地理位置坐标对应的多张历史路况实景图片，并获取每一历史路况实景图片的采集时间；Obtain a plurality of historical road condition real-scene pictures corresponding to the geographic location coordinates, and obtain the collection time of each historical road condition real-scene picture;
获取多张历史路况实景图片对应的采集时间与当前时刻的时间间距为最小时间间隔、且对应的道路天气状况识别结果为非恶劣天气状况的历史路况实景图片。Acquiring multiple historical road condition real pictures corresponding to the acquisition time and the current moment with a minimum time interval, and the corresponding road weather condition recognition results are historical road condition real pictures that are not severe weather conditions.
其中，非恶劣天气状况是晴天或阴天，在这两种天气情况下路面还是清晰可见的，故道路在非恶劣天气状况下的道路路况实景图片替换为该道路在非恶劣天气状况下的道路路况实景图片，能有效通过该道路在非恶劣天气状况下的道路路况实景图片来作为导航图片。Among them, the non-bad weather conditions are sunny or cloudy, and the road surface is still clearly visible under these two weather conditions, so the real picture of the road conditions of the road in non-bad weather conditions is replaced by the road of the road in non-bad weather conditions The real picture of the road condition can effectively be used as a navigation picture by using the real picture of the road condition of the road under non-bad weather conditions.
若该地理位置的实景图片有多种时，优先选择距离当前系统时间最近的一次采集所得到的实景图片进行替换；例如，距离当前系统时间最近的一次采集为30分钟之前，而且30分钟之前为非恶劣天气，此时可以获取30分钟之前采集的道路路况实景图片来作为导航图片；距离当前系统时间最近的一次采集为30分钟之前，而且30分钟之前为恶劣天气，此时可以获取1个小时之前采集的道路路况实景图片再次判断是否为恶劣天气，直至向前推的某一时刻所采集的道路路况实景图片对应为非恶劣天气。If there are multiple real-scene pictures in this geographical location, the real-scene picture obtained from the closest collection to the current system time is preferred for replacement; If it is not bad weather, you can get the real picture of road conditions collected 30 minutes ago as a navigation picture; the latest collection from the current system time was 30 minutes ago, and it was bad weather 30 minutes ago, you can get 1 hour at this time It is judged again whether the real picture of the road condition collected before is bad weather, until the real picture of the road condition collected at a certain moment forward corresponds to non-bad weather.
该方法实现了按周期获取道路路况图片，并且及时更新到电子地图上，基于实时路况数据，在复杂路面情况下时可有效导航。The method realizes periodic acquisition of road condition pictures and updates them on the electronic map in time, and based on the real-time traffic condition data, it can effectively navigate under complex road conditions.
本发明实施例还提供一种基于历史视频的导航装置，该基于历史视频的导航装置用于执行前述基于历史视频的导航方法的任一实施例。具体地，请参阅图4，图4是本发明实施例提供的基于历史视频的导航装置的示意性框图。该基于历史视频的导航装置100可以配置于服务器中。An embodiment of the present invention also provides a historical video-based navigation device, which is used to execute any embodiment of the aforementioned historical video-based navigation method. Specifically, please refer to FIG. 4 , which is a schematic block diagram of a historical video-based navigation device according to an embodiment of the present invention. The historical video-based navigation device 100 can be configured in a server.
如图4所示，基于历史视频的导航装置100包括当前视频采集单元110、当前视频分解单元120、随机选取单元130、天气状况识别单元140、结果判断单元150、图片替换单元160。As shown in FIG. 4 , the historical video-based navigation device 100 includes a current video acquisition unit 110 , a current video decomposition unit 120 , a random selection unit 130 , a weather condition identification unit 140 , a result judgment unit 150 , and a picture replacement unit 160 .
当前视频采集单元110，用于若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据。The current video acquisition unit 110 is configured to acquire road condition video data of the road corresponding to the preset geographical location coordinates at the current moment if the time interval between the current moment and the previous video acquisition moment is equal to the preset acquisition period.
在本实施例中，对设置在道路路口的前端采集装置(如高清摄像头)设置一个采集周期，是为了周期性的对各被监控的道路进行路况视频记录，每间隔采集周期(例如设置采集周期为30分钟)所采集的道路路况视频数据的时长也是相同的(例如5秒钟)，并可以对所采集的道路路况视频数据进行处理后，得到最新路况的导航图片。In this embodiment, a collection cycle is set for the front-end collection device (such as a high-definition camera) arranged at a road intersection, in order to periodically record the road condition video of each monitored road, and every interval collection cycle (such as setting the collection cycle The duration of the road traffic video data collected for 30 minutes) is also the same (for example, 5 seconds), and after the collected road traffic video data can be processed, a navigation picture of the latest road conditions can be obtained.
当前视频分解单元120，用于将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片。The current video decomposition unit 120 is configured to perform video decomposition on the road traffic condition video data to obtain multiple frames of road traffic condition pictures corresponding to the road traffic condition video data.
在本实施例中，由视频的形成过程可知，视频是由多帧图片组成，例如每一秒的视频能转化成24-30张图片。通过视频分解工具(如Adobe Premiere，在Adobe Premiere中剪辑出需要输出序列位图的片段，从文件菜单里输出成序列位图即可)即可将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片。In this embodiment, it can be seen from the video forming process that the video is composed of multiple frames of pictures, for example, every second of video can be converted into 24-30 pictures. Through the video decomposition tool (such as Adobe Premiere, in Adobe Premiere, edit the segment that needs to output the sequence bitmap, and output it as a sequence bitmap from the file menu) the video decomposition of the road traffic condition video data can be carried out to obtain the same Multiple frames of road traffic picture corresponding to the road traffic video data.
随机选取单元130，用于在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片。The random selection unit 130 is configured to randomly acquire a frame of pictures from the multi-frame road traffic picture corresponding to the road traffic video data as an actual road traffic picture corresponding to the road traffic video data.
在本实施例中，在当前时刻所采集预设时长(5秒钟)的道路路况视频数据，进行视频分解后，得到了120-150张道路路况图片，此时可以在其中随机选择一张作为当前时刻的道路路况图片，这样针对每一路段每间隔采集周期采集到道路路况视频数据仅保留一张图片作为道路路况实景图片。In this embodiment, the road traffic condition video data of the preset duration (5 seconds) collected at the current moment is decomposed into 120-150 road traffic condition pictures, and one of them can be randomly selected as The road condition picture at the current moment, so that only one picture is reserved for the road condition video data collected at each interval collection period for each road section as the real picture of the road condition.
在一实施例中，基于历史视频的导航装置100还包括：In one embodiment, the historical video-based navigation device 100 also includes:
图片定时存储单元，用于将当前时刻的实际道路路况实景图片及对应的采集时间进行存储。The picture timing storage unit is used to store the actual picture of the actual road condition at the current moment and the corresponding collection time.
在具体实施时，针对每一个被监控的道路，在每间隔所述采集周期所采集的道路路况视频数据经视频分解和随机挑选一帧图片后仅保留一张道路路况实景图片，则每一被监控的道路在多个道路路况视频采集周期所积累下来有多张道路路况实景图片。例如道路路况视频采集周期设置为30分钟，则针对每一被监控的道路一天有48张道路路况实景图片进行保存。一旦该道路的状态发生改变，均可及时的上传提供导航服务的服务器，由服务器及时的更新各被监控的道路的道路路况实景图片。During specific implementation, for each monitored road, only one road traffic real scene picture is reserved after video decomposition and random selection of a frame of the road traffic video data collected at each interval of the collection period, then each monitored road The monitored road has multiple real-life pictures of road conditions accumulated in multiple road condition video collection cycles. For example, if the collection cycle of the road condition video is set to 30 minutes, then 48 real pictures of the road condition will be saved for each monitored road a day. Once the state of the road changes, it can be uploaded in time to the server providing the navigation service, and the real picture of the road condition of each monitored road can be updated in time by the server.
通过这种周期性更新道路的路况实景图片，能较为及时的提醒使用导航服务的用户每一道路的实际道路情况。例如，某一时刻当前道路进入临时维护状态，此时刻对应获取的该道路的道路路况实景图片中能及时体现该道路进入了维修状态，可能导致拥堵，提示司机及时规划其他路线绕行。By periodically updating the road condition real picture of the road, the user using the navigation service can be reminded of the actual road condition of each road in a timely manner. For example, if the current road enters a temporary maintenance state at a certain moment, the road condition real picture corresponding to the acquired road at this moment can reflect that the road has entered the maintenance state in time, which may lead to congestion, prompting the driver to plan other detours in time.
天气状况识别单元140，用于将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果。The weather condition recognition unit 140 is configured to use the real road condition picture as an input of a pre-trained road weather condition recognition model to obtain a road weather condition recognition result corresponding to the real road condition picture.
在一实施例中，基于历史视频的导航装置100还包括：In one embodiment, the historical video-based navigation device 100 also includes:
模型训练单元，用于获取多个训练数据，以对待训练的道路天气状况识别模型进行训练，得到用于识别道路天气状况识别结果的道路天气状况识别模型；其中，多个训练数据中每一训练数据以道路路况图片对应的图片特征向量作为待训练的道路天气状况识别模型的输入，并以道路天气状况识别结果作为待训练的道路天气状况识别模型的输出；其中，多个训练数据中每一训练数据对应的道路路况图片的采集时刻与当前时刻之差小于或等于预设的时间间隔阈值；每一训练数据对应的道路天气状况识别结果为预先标注的天气状况识别结果。The model training unit is used to obtain a plurality of training data to train the road weather condition recognition model to be trained to obtain a road weather condition recognition model for recognizing the result of road weather condition recognition; wherein, each of the plurality of training data is trained The data uses the picture feature vector corresponding to the road condition picture as the input of the road weather condition recognition model to be trained, and uses the road weather condition recognition result as the output of the road weather condition recognition model to be trained; wherein, each of a plurality of training data The difference between the collection time of the road condition picture corresponding to the training data and the current time is less than or equal to the preset time interval threshold; the road weather condition recognition result corresponding to each training data is the pre-marked weather condition recognition result.
在本实施例中，预先训练一个道路天气状况识别模型时，可以选定当前道路多张天气状况下所拍摄的道路路况实景图片，如在雨天、下雪天、雾霾天、晴天、阴天5种天气状况下的道路路况实景图片，然后设置晴天对应的道路天气状况识别结果为1，阴天对应的道路天气状况识别结果为2，雨天对应的道路天气状况识别结果为3，下雪天对应的道路天气状况识别结果为4，雾霾天对应的道路天气状况识别结果为5。通过大量的各种天气状况下的道路路况实景图片对应的图片特征向量作为输入，且将对应设置的道路天气状况识别结果作为输出，对深度神经网络进训练得到道路天气状况识别模型。In this embodiment, when pre-training a road weather condition recognition model, the real picture of the road condition taken under the current road condition can be selected, such as rainy day, snowy day, haze day, sunny day, cloudy day Real-scene pictures of road conditions under 5 kinds of weather conditions, and then set the road weather condition recognition result corresponding to sunny days as 1, the road weather condition recognition result corresponding to cloudy days as 2, the road weather condition recognition result corresponding to rainy days as 3, and the road weather condition recognition result corresponding to snowy days as 3. The corresponding road weather condition identification result is 4, and the road weather condition identification result corresponding to the haze day is 5. Through a large number of picture feature vectors corresponding to real road conditions under various weather conditions as input, and the corresponding set road weather condition recognition results as output, the deep neural network is trained to obtain the road weather condition recognition model.
所采用的深度神经网络即DNN(Deep Neural Networks，简称DNN)，DNN可以理解为有很多隐藏层的神经网络。DNN内部的神经网络层可以分为三类，输入层，隐藏层和输出层,一般来说第一层是输入层，最后一层是输出层，而中间的层数都是隐藏层。层与层之间是全连接的，也就是说，第i层的任意一个神经元一定与第i+1层的任意一个神经元相连。The deep neural network used is DNN (Deep Neural Networks, DNN for short), and DNN can be understood as a neural network with many hidden layers. The neural network layers inside the DNN can be divided into three categories, the input layer, the hidden layer and the output layer. Generally speaking, the first layer is the input layer, the last layer is the output layer, and the middle layers are all hidden layers. The layers are fully connected, that is, any neuron in the i-th layer must be connected to any neuron in the i+1-th layer.
在一实施例中，如图5所示，天气状况识别单元140包括：In one embodiment, as shown in FIG. 5 , the weather condition identification unit 140 includes:
像素矩阵获取单元141，用于获取与所述实际道路路况实景图片对应的像素矩阵；A pixel matrix acquisition unit 141, configured to acquire a pixel matrix corresponding to the real picture of the actual road condition;
卷积单元142，用于将所述像素矩阵作为卷积神经网络模型中输入层的输入，对应得到多个特征图；The convolution unit 142 is used to use the pixel matrix as the input of the input layer in the convolutional neural network model to obtain multiple feature maps correspondingly;
池化单元143，用于将多个特征图输入至卷积神经网络模型的池化层，得到每一特征图对应的最大值所对应一维行向量；The pooling unit 143 is used to input a plurality of feature maps to the pooling layer of the convolutional neural network model to obtain a one-dimensional row vector corresponding to the maximum value corresponding to each feature map;
全连接单元144，用于将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量；The fully connected unit 144 is used to input the one-dimensional row vector corresponding to the maximum value corresponding to each feature map to the fully connected layer of the convolutional neural network model, to obtain the picture feature vector corresponding to the real picture of the actual road condition;
输入向量获取单元145，用于将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。The input vector obtaining unit 145 is configured to use the picture feature vector as an input of a pre-trained road weather condition recognition model.
在本实施例中，每一帧道路路况实景图片均可以转化为像素矩阵(如256*256的像素矩阵)，将像素矩阵作为卷积神经网络模型中输入层的输入时，输入层通过卷积操作得到若干个Feature Map(Feature Map可以理解为特征图)，在设置好卷积窗口的大小后，通过卷积窗口对像素矩阵进行卷积将得到若干个列数为1的Feature Map。In this embodiment, each frame of road conditions real picture can be converted into a pixel matrix (such as a 256*256 pixel matrix), when the pixel matrix is used as the input of the input layer in the convolutional neural network model, the input layer passes through The operation obtains several Feature Maps (Feature Maps can be understood as feature maps). After setting the size of the convolution window, the pixel matrix is convoluted through the convolution window to obtain several Feature Maps with a column number of 1.
在池化层中，可采用从之前一维的Feature Map中提出最大的值。通过这种Pooling方式(即池化的方式)可以解决可变图片像素大小的输入问题(因为不管FeatureMap中有多少个值，只需要提取其中的最大值)，最终池化层的输出为各个Feature Map的最大值，即一个一维的向量。In the pooling layer, the largest value can be proposed from the previous one-dimensional Feature Map. Through this Pooling method (that is, pooling method), the input problem of variable image pixel size can be solved (because no matter how many values there are in FeatureMap, only the maximum value needs to be extracted), and the output of the final pooling layer is each Feature The maximum value of Map, that is, a one-dimensional vector.
最后将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量，即可将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。Finally, the one-dimensional row vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer of the convolutional neural network model, and the picture feature vector corresponding to the real picture of the actual road condition is obtained, and the picture feature can be vector as input to a pre-trained road weather condition recognition model.
结果判断单元150，用于若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标。The result judging unit 150 is configured to acquire a corresponding actual road condition real picture and the corresponding geographic position coordinates of the actual road condition real picture if the road weather condition recognition result is greater than a preset severe weather warning value.
在本实施例中，当通过道路天气状况识别模型对所述实际道路路况实景图片进行识别时，若当前时刻所采集的道路路况实景图片对应的道路天气状况识别结果为3，超出了预设的恶劣天气预警值2，则表示当前道路此时刻的道路路况实景图片不适宜作为道路的实景导航图片。此时需定位存在道路天气状况识别结果大于预设的恶劣天气预警值所对应道路的地理位置坐标以判断是哪条道路的实景导航图片需要替换为非恶劣天气状况下道路路况实景图片(如晴天或阴天下的道路路况实景图片)，一般选择与当前时间间隔最近的非恶劣天气状况下道路路况实景图片进行替换。In this embodiment, when the road weather condition identification model is used to identify the actual road condition real picture, if the road weather condition identification result corresponding to the road condition real picture collected at the current moment is 3, exceeding the preset A severe weather warning value of 2 indicates that the real picture of the current road condition at this moment is not suitable as a real picture of the road for navigation. At this time, it is necessary to locate the geographic location coordinates of the road corresponding to the road weather condition identification result greater than the preset severe weather warning value to determine which road the real-scene navigation picture needs to be replaced with a real-scene picture of the road condition in non-severe weather conditions (such as a sunny day) or a real picture of the road condition under cloudy sky), generally select the real picture of the road condition under the non-bad weather condition closest to the current time interval to replace.
图片替换单元160，用于将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。The picture replacement unit 160 is configured to replace the real picture of the actual road condition corresponding to the geographic location coordinates with the real picture of the road condition under non-bad weather conditions.
在一实施例中，于历史视频的导航装置100还包括：In one embodiment, the historical video navigation device 100 also includes:
历史路况实景图片获取单元，用于获取与所述地理位置坐标对应的多张历史路况实景图片，并获取每一历史路况实景图片的采集时间；A historical road condition real picture acquisition unit, configured to acquire a plurality of historical road condition real pictures corresponding to the geographic location coordinates, and acquire the collection time of each historical road condition real picture;
图片筛选单元，用于获取多张历史路况实景图片对应的采集时间与当前时刻的时间间距为最小时间间隔、且对应的道路天气状况识别结果为非恶劣天气状况的历史路况实景图片。The picture screening unit is used to acquire multiple historical road condition real pictures corresponding to the collection time and the current moment at a minimum time interval, and the corresponding road weather condition recognition result is a historical road condition real picture that is not a bad weather condition.
其中，非恶劣天气状况是晴天或阴天，在这两种天气情况下路面还是清晰可见的，故道路在非恶劣天气状况下的道路路况实景图片替换为该道路在非恶劣天气状况下的道路路况实景图片，能有效通过该道路在非恶劣天气状况下的道路路况实景图片来作为导航图片。Among them, the non-bad weather conditions are sunny or cloudy, and the road surface is still clearly visible under these two weather conditions, so the real picture of the road conditions of the road in non-bad weather conditions is replaced by the road of the road in non-bad weather conditions The real picture of the road condition can effectively be used as a navigation picture by using the real picture of the road condition of the road under non-bad weather conditions.
若该地理位置的实景图片有多种时，优先选择距离当前系统时间最近的一次采集所得到的实景图片进行替换；例如，距离当前系统时间最近的一次采集为30分钟之前，而且30分钟之前为非恶劣天气，此时可以获取30分钟之前采集的道路路况实景图片来作为导航图片；距离当前系统时间最近的一次采集为30分钟之前，而且30分钟之前为恶劣天气，此时可以获取1个小时之前采集的道路路况实景图片再次判断是否为恶劣天气，直至向前推的某一时刻所采集的道路路况实景图片对应为非恶劣天气。If there are multiple real-scene pictures in this geographical location, the real-scene picture obtained from the closest collection to the current system time is preferred for replacement; If it is not bad weather, you can get the real picture of road conditions collected 30 minutes ago as a navigation picture; the latest collection from the current system time was 30 minutes ago, and it was bad weather 30 minutes ago, you can get 1 hour at this time It is judged again whether the real picture of the road condition collected before is bad weather, until the real picture of the road condition collected at a certain moment forward corresponds to non-bad weather.
该装置实现了按周期获取道路路况图片，并且及时更新到电子地图上，基于实时路况数据，在复杂路面情况下时可有效导航。The device realizes periodic acquisition of road condition pictures and updates them on the electronic map in time, based on real-time traffic data, it can effectively navigate under complex road conditions.
上述基于历史视频的导航装置可以实现为计算机程序的形式，该计算机程序可以在如图6所示的计算机设备上运行。The above-mentioned navigation device based on historical video can be realized in the form of a computer program, and the computer program can be run on the computer equipment as shown in FIG. 6 .
请参阅图6，图6是本发明实施例提供的计算机设备的示意性框图。该计算机设备500是服务器，服务器可以是独立的服务器，也可以是多个服务器组成的服务器集群。Please refer to FIG. 6. FIG. 6 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device 500 is a server, and the server may be an independent server or a server cluster composed of multiple servers.
参阅图6，该计算机设备500包括通过系统总线501连接的处理器502、存储器和网络接口505，其中，存储器可以包括非易失性存储介质503和内存储器504。Referring to FIG. 6 , the computer device 500 includes a processor 502 connected through a system bus 501 , a memory and a network interface 505 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .
该非易失性存储介质503可存储操作系统5031和计算机程序5032。该计算机程序5032被执行时，可使得处理器502执行基于历史视频的导航方法。The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032 . When the computer program 5032 is executed, it can cause the processor 502 to execute the historical video-based navigation method.
该处理器502用于提供计算和控制能力，支撑整个计算机设备500的运行。The processor 502 is used to provide calculation and control capabilities and support the operation of the entire computer device 500 .
该内存储器504为非易失性存储介质503中的计算机程序5032的运行提供环境，该计算机程序5032被处理器502执行时，可使得处理器502执行基于历史视频的导航方法。The internal memory 504 provides an environment for running the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the historical video-based navigation method.
该网络接口505用于进行网络通信，如提供数据信息的传输等。本领域技术人员可以理解，图6中示出的结构，仅仅是与本发明方案相关的部分结构的框图，并不构成对本发明方案所应用于其上的计算机设备500的限定，具体的计算机设备500可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。The network interface 505 is used for network communication, such as providing data transmission and the like. Those skilled in the art can understand that the structure shown in FIG. 6 is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation to the computer device 500 on which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than shown, or combine certain components, or have a different arrangement of components.
其中，所述处理器502用于运行存储在存储器中的计算机程序5032，以实现如下功能：若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据；将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片；在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片；将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果；若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标；以及将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。Wherein, the processor 502 is used to run the computer program 5032 stored in the memory to realize the following functions: if the time interval between the current moment and the previous video acquisition moment is equal to the preset acquisition cycle, obtain the preset geographic location coordinates The road traffic video data of the corresponding road at the current moment; the video decomposition of the road traffic video data is carried out to obtain a multi-frame road traffic picture corresponding to the road traffic video data; Randomly obtain a frame of pictures in the multi-frame road condition pictures as the actual road condition real scene picture corresponding to the road condition video data; use the actual road condition real scene picture as the input of the pre-trained road weather condition recognition model to obtain and the road weather condition recognition result of the actual road condition real picture; if the road weather condition recognition result is greater than the preset severe weather warning value, obtain the corresponding actual road condition real picture, corresponding to the actual road condition real picture The geographic location coordinates of the location coordinates; and the actual road condition real picture corresponding to the geographic location coordinate is replaced by the road traffic real scene picture under non-bad weather conditions.
在一实施例中，处理器502在执行所述获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据的步骤之前，还执行如下操作：获取多个训练数据，以对待训练的道路天气状况识别模型进行训练，得到用于识别道路天气状况识别结果的道路天气状况识别模型；其中，多个训练数据中每一训练数据以道路路况图片对应的图片特征向量作为待训练的道路天气状况识别模型的输入，并以道路天气状况识别结果作为待训练的道路天气状况识别模型的输出。In an embodiment, before the processor 502 executes the step of obtaining the road traffic condition video data of the road corresponding to the preset geographic location coordinates at the current moment, the following operation is further performed: obtaining a plurality of training data for training The road weather condition identification model is trained to obtain a road weather condition identification model for identifying the road weather condition identification result; wherein, each training data in a plurality of training data uses the picture feature vector corresponding to the road condition picture as the road to be trained The input of the weather condition recognition model, and the road weather condition recognition result is used as the output of the road weather condition recognition model to be trained.
在一实施例中，处理器502在执行所述将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入的步骤时，执行如下操作：获取与所述实际道路路况实景图片对应的像素矩阵；将所述像素矩阵作为卷积神经网络模型中输入层的输入，对应得到多个特征图；将多个特征图输入至卷积神经网络模型的池化层，得到每一特征图对应的最大值所对应一维行向量；将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量；将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。In one embodiment, when the processor 502 executes the step of using the real road condition picture as the input of the pre-trained road weather condition recognition model, the following operation is performed: obtain the real picture corresponding to the actual road condition The pixel matrix; the pixel matrix is used as the input of the input layer in the convolutional neural network model, and multiple feature maps are correspondingly obtained; multiple feature maps are input to the pooling layer of the convolutional neural network model to obtain each feature map The one-dimensional row vector corresponding to the corresponding maximum value; the one-dimensional row vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer of the convolutional neural network model, and the picture corresponding to the real picture of the actual road condition is obtained A feature vector; the picture feature vector is used as an input of a pre-trained road weather condition recognition model.
在一实施例中，处理器502在执行所述所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片的步骤之前，还执行如下操作：获取与所述地理位置坐标对应的多张历史路况实景图片，并获取每一历史路况实景图片的采集时间；获取多张历史路况实景图片对应的采集时间与当前时刻的时间间距为最小时间间隔、且对应的道路天气状况识别结果为非恶劣天气状况的历史路况实景图片。In one embodiment, before the processor 502 executes the step of replacing the actual road condition real picture corresponding to the geographic location coordinates with the road condition real picture under non-bad weather conditions, the following operations are further performed: acquiring the Multiple historical road condition real pictures corresponding to geographic location coordinates, and obtain the collection time of each historical road condition real picture; the time interval between the collection time corresponding to multiple historical road condition real pictures and the current moment is the minimum time interval, and the corresponding road The weather condition recognition result is a real picture of the historical road condition that is not severe weather condition.
在一实施例中，处理器502在执行所述在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片的步骤之后，还执行如下操作：将当前时刻的实际道路路况实景图片及对应的采集时间进行存储。In one embodiment, the processor 502 randomly acquires a frame of pictures from the multi-frame road traffic picture corresponding to the road traffic video data as the actual road traffic scene corresponding to the road traffic video data. After the step of taking pictures, the following operation is also performed: storing the real picture of the actual road condition at the current moment and the corresponding collection time.
本领域技术人员可以理解，图6中示出的计算机设备的实施例并不构成对计算机设备具体构成的限定，在其他实施例中，计算机设备可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。例如，在一些实施例中，计算机设备可以仅包括存储器及处理器，在这样的实施例中，存储器及处理器的结构及功能与图6所示实施例一致，在此不再赘述。Those skilled in the art can understand that the embodiment of the computer device shown in FIG. 6 does not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or less components than those shown in the illustration. Or combine certain components, or different component arrangements. For example, in some embodiments, the computer device may only include a memory and a processor. In such an embodiment, the structures and functions of the memory and the processor are consistent with those of the embodiment shown in FIG. 6 , and will not be repeated here.
应当理解，在本发明实施例中，处理器502可以是中央处理单元(CentralProcessing Unit，CPU)，该处理器502还可以是其他通用处理器、数字信号处理器(DigitalSignal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable GateArray，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。其中，通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。It should be understood that, in the embodiment of the present invention, the processor 502 may be a central processing unit (Central Processing Unit, CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), dedicated integrated Circuit (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate array (Field-Programmable GateArray, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Wherein, the general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
在本发明的另一实施例中提供计算机可读存储介质。该计算机可读存储介质可以为非易失性的计算机可读存储介质。该计算机可读存储介质存储有计算机程序，其中计算机程序被处理器执行时实现以下步骤：若当前时刻与前一视频采集时刻的时间间隔等于预设的采集周期，获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据；将所述道路路况视频数据进行视频分解，得到与所述道路路况视频数据对应的多帧道路路况图片；在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片；将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，得到与所述实际道路路况实景图片的道路天气状况识别结果；若所述道路天气状况识别结果大于预设的恶劣天气预警值，获取对应的实际道路路况实景图片，及所述实际道路路况实景图片对应的地理位置坐标；以及将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片。In another embodiment of the invention a computer readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by the processor, the following steps are implemented: if the time interval between the current moment and the previous video acquisition moment is equal to the preset acquisition cycle, obtain the preset geographic location coordinates The road traffic condition video data of the corresponding road at the current moment; the video decomposition of the road traffic condition video data is carried out to obtain a multi-frame road traffic condition picture corresponding to the road traffic condition video data; in the multiple frames corresponding to the road traffic condition video data Randomly obtain a frame of pictures in the frame road condition picture, as the actual road condition real-scene picture corresponding with described road condition video data; With described actual road condition real-scene picture as the input of the road weather condition identification model of pre-training, obtain and The road weather condition identification result of the actual road condition real picture; if the road weather condition identification result is greater than the preset severe weather warning value, obtain the corresponding actual road condition real picture, and the corresponding actual road condition real picture Geographic location coordinates; and replacing the actual road condition real picture corresponding to the geographical location coordinate with the road condition real picture under non-bad weather conditions.
在一实施例中，所述获取预设的地理位置坐标所对应的道路在当前时刻的道路路况视频数据之前，还包括：获取多个训练数据，以对待训练的道路天气状况识别模型进行训练，得到用于识别道路天气状况识别结果的道路天气状况识别模型；其中，多个训练数据中每一训练数据以道路路况图片对应的图片特征向量作为待训练的道路天气状况识别模型的输入，并以道路天气状况识别结果作为待训练的道路天气状况识别模型的输出。In one embodiment, the obtaining of the road corresponding to the preset geographic location coordinates before the road traffic video data at the current moment further includes: obtaining a plurality of training data to train the road weather condition recognition model to be trained, Obtain the road weather condition recognition model that is used to identify the road weather condition recognition result; Wherein, in a plurality of training data, each training data uses the image feature vector corresponding to the road condition picture as the input of the road weather condition recognition model to be trained, and with The road weather condition recognition result is used as the output of the road weather condition recognition model to be trained.
在一实施例中，所述将所述实际道路路况实景图片作为预先训练的道路天气状况识别模型的输入，包括：获取与所述实际道路路况实景图片对应的像素矩阵；将所述像素矩阵作为卷积神经网络模型中输入层的输入，对应得到多个特征图；将多个特征图输入至卷积神经网络模型的池化层，得到每一特征图对应的最大值所对应一维行向量；将每一特征图对应的最大值所对应一维行向量输入至卷积神经网络模型的全连接层，得到与所述实际道路路况实景图片对应的图片特征向量；将所述图片特征向量作为预先训练的道路天气状况识别模型的输入。In one embodiment, said using the actual road condition picture as the input of the pre-trained road weather condition recognition model includes: acquiring a pixel matrix corresponding to the actual road condition picture; using the pixel matrix as The input of the input layer in the convolutional neural network model corresponds to multiple feature maps; input multiple feature maps to the pooling layer of the convolutional neural network model to obtain the one-dimensional row vector corresponding to the maximum value of each feature map ; The one-dimensional row vector corresponding to the maximum value corresponding to each feature map is input to the fully connected layer of the convolutional neural network model, and the picture feature vector corresponding to the actual road condition real scene picture is obtained; the picture feature vector is used as Input to a pre-trained road weather condition recognition model.
在一实施例中，所述将所述地理位置坐标对应的实际道路路况实景图片替换为非恶劣天气状况下的道路路况实景图片之前，还包括：获取与所述地理位置坐标对应的多张历史路况实景图片，并获取每一历史路况实景图片的采集时间；获取多张历史路况实景图片对应的采集时间与当前时刻的时间间距为最小时间间隔、且对应的道路天气状况识别结果为非恶劣天气状况的历史路况实景图片。In one embodiment, before replacing the actual road condition picture corresponding to the geographic location coordinates with the road condition real picture under non-bad weather conditions, it also includes: obtaining multiple historical records corresponding to the geographic location coordinates Real-scene pictures of road conditions, and obtain the collection time of each historical real-scene picture of road conditions; the time interval between the collection time and the current moment corresponding to multiple historical road-condition pictures is the minimum time interval, and the corresponding road weather condition recognition result is non-bad weather Real-life pictures of historical road conditions.
在一实施例中，所述在与所述道路路况视频数据对应的多帧道路路况图片中随机获取一帧图片，以作为与所述道路路况视频数据对应的实际道路路况实景图片之后，包括：将当前时刻的实际道路路况实景图片及对应的采集时间进行存储。In one embodiment, after randomly acquiring a frame of pictures from the multi-frame road traffic picture corresponding to the road traffic video data as the actual road traffic scene picture corresponding to the road traffic video data, it includes: Store the real picture of the actual road condition at the current moment and the corresponding collection time.
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，上述描述的设备、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the foregoing method embodiments, and details are not repeated here. Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the relationship between hardware and software Interchangeability. In the above description, the composition and steps of each example have been generally described according to their functions. Whether these functions are implemented by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present invention.
在本发明所提供的几个实施例中，应该理解到，所揭露的设备、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为逻辑功能划分，实际实现时可以有另外的划分方式，也可以将具有相同功能的单元集合成一个单元，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口、装置或单元的间接耦合或通信连接，也可以是电的，机械的或其它的形式连接。In the several embodiments provided by the present invention, it should be understood that the disclosed devices, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only logical function division. In actual implementation, there may be other division methods, and units with the same function may also be combined into one Units such as multiple units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may also be electrical, mechanical or other forms of connection.
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本发明实施例方案的目的。The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present invention.
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以是两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分，或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、磁碟或者光盘等各种可以存储程序代码的介质。If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of software products, and the computer software products are stored in a storage medium In, several instructions are included to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media capable of storing program codes such as a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a magnetic disk, or an optical disk.
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到各种等效的修改或替换，这些修改或替换都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以权利要求的保护范围为准。The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technical field can easily think of various equivalents within the technical scope disclosed in the present invention. Modifications or replacements shall all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910371108.4ACN110186471A (en) | 2019-05-06 | 2019-05-06 | Air navigation aid, device, computer equipment and storage medium based on history video |
| PCT/CN2019/103329WO2020224117A1 (en) | 2019-05-06 | 2019-08-29 | Historical video-based navigation method and device, computer apparatus and storage medium |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910371108.4ACN110186471A (en) | 2019-05-06 | 2019-05-06 | Air navigation aid, device, computer equipment and storage medium based on history video |
| Publication Number | Publication Date |
|---|---|
| CN110186471Atrue CN110186471A (en) | 2019-08-30 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910371108.4APendingCN110186471A (en) | 2019-05-06 | 2019-05-06 | Air navigation aid, device, computer equipment and storage medium based on history video |
| Country | Link |
|---|---|
| CN (1) | CN110186471A (en) |
| WO (1) | WO2020224117A1 (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807549A (en)* | 2019-10-29 | 2020-02-18 | 国网电力科学研究院武汉南瑞有限责任公司 | Generation method, generation device, generation system and electronic equipment of meteorological prediction model |
| CN111178407A (en)* | 2019-12-19 | 2020-05-19 | 中国平安人寿保险股份有限公司 | Road condition data screening method and device, computer equipment and storage medium |
| CN113887286A (en)* | 2021-08-31 | 2022-01-04 | 际络科技(上海)有限公司 | Driver behavior monitoring method based on online video understanding network |
| CN115148040A (en)* | 2022-06-28 | 2022-10-04 | 东莞中科云计算研究院 | Unmanned vehicle control method and system for closed road environment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008257403A (en)* | 2007-04-03 | 2008-10-23 | Denso Corp | Driving support device for vehicle |
| JP2010257253A (en)* | 2009-04-24 | 2010-11-11 | Denso Corp | Vehicle display device |
| CN104834912A (en)* | 2015-05-14 | 2015-08-12 | 北京邮电大学 | Weather identification method and apparatus based on image information detection |
| CN105575119A (en)* | 2015-12-29 | 2016-05-11 | 大连楼兰科技股份有限公司 | Method and device for deep learning and recognition of road conditions and climate |
| CN106611162A (en)* | 2016-12-20 | 2017-05-03 | 西安电子科技大学 | Method for real-time detection of road vehicle based on deep learning SSD frame |
| CN106709511A (en)* | 2016-12-08 | 2017-05-24 | 华中师范大学 | Urban rail transit panoramic monitoring video fault detection method based on depth learning |
| CN107238874A (en)* | 2017-05-25 | 2017-10-10 | 苏州工业职业技术学院 | Based on when scape photo real-time weather detection method and system |
| EP3361412A1 (en)* | 2017-02-10 | 2018-08-15 | Fujitsu Limited | Black ice detection system, program, and method |
| CN108513674A (en)* | 2018-03-26 | 2018-09-07 | 深圳市锐明技术股份有限公司 | A kind of detection alarm method, storage medium and the server of Chinese herbaceous peony accumulated snow and icing |
| CN108701396A (en)* | 2018-03-26 | 2018-10-23 | 深圳市锐明技术股份有限公司 | A kind of detection alarm method, storage medium and the server of Chinese herbaceous peony accumulated snow and icing |
| CN108805198A (en)* | 2018-06-08 | 2018-11-13 | Oppo广东移动通信有限公司 | Image processing method, device, computer readable storage medium and electronic equipment |
| CN108875593A (en)* | 2018-05-28 | 2018-11-23 | 上海交通大学 | Visible images weather recognition methods based on convolutional neural networks |
| CN108921013A (en)* | 2018-05-16 | 2018-11-30 | 浙江零跑科技有限公司 | A kind of visual scene identifying system and method based on deep neural network |
| CN109447018A (en)* | 2018-11-08 | 2019-03-08 | 天津理工大学 | A kind of road environment visual perception method based on improvement Faster R-CNN |
| CN109525901A (en)* | 2018-11-27 | 2019-03-26 | Oppo广东移动通信有限公司 | Method for processing video frequency, device, electronic equipment and computer-readable medium |
| DE102017217072A1 (en)* | 2017-09-26 | 2019-03-28 | Volkswagen Aktiengesellschaft | A method for detecting a weather condition in an environment of a motor vehicle and control device and motor vehicle |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105628034B (en)* | 2016-02-04 | 2019-04-23 | 合肥杰发科技有限公司 | Navigation map update method and device |
| WO2018225446A1 (en)* | 2017-06-09 | 2018-12-13 | 株式会社デンソー | Map points-of-change detection device |
| JP7062892B2 (en)* | 2017-07-13 | 2022-05-09 | トヨタ自動車株式会社 | Dynamic map updater, dynamic map update method, dynamic map update program |
| CN108196285B (en)* | 2017-11-30 | 2021-12-17 | 中山大学 | Accurate positioning system based on multi-sensor fusion |
| CN108282635B (en)* | 2018-02-11 | 2020-11-10 | 中国联合网络通信集团有限公司 | Panoramic image generation method and system, Internet of vehicles big data service platform |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008257403A (en)* | 2007-04-03 | 2008-10-23 | Denso Corp | Driving support device for vehicle |
| JP2010257253A (en)* | 2009-04-24 | 2010-11-11 | Denso Corp | Vehicle display device |
| CN104834912A (en)* | 2015-05-14 | 2015-08-12 | 北京邮电大学 | Weather identification method and apparatus based on image information detection |
| US20160334546A1 (en)* | 2015-05-14 | 2016-11-17 | Beijing University Of Posts And Telecommunications | Weather recognition method and device based on image information detection |
| CN105575119A (en)* | 2015-12-29 | 2016-05-11 | 大连楼兰科技股份有限公司 | Method and device for deep learning and recognition of road conditions and climate |
| CN106709511A (en)* | 2016-12-08 | 2017-05-24 | 华中师范大学 | Urban rail transit panoramic monitoring video fault detection method based on depth learning |
| CN106611162A (en)* | 2016-12-20 | 2017-05-03 | 西安电子科技大学 | Method for real-time detection of road vehicle based on deep learning SSD frame |
| EP3361412A1 (en)* | 2017-02-10 | 2018-08-15 | Fujitsu Limited | Black ice detection system, program, and method |
| CN107238874A (en)* | 2017-05-25 | 2017-10-10 | 苏州工业职业技术学院 | Based on when scape photo real-time weather detection method and system |
| DE102017217072A1 (en)* | 2017-09-26 | 2019-03-28 | Volkswagen Aktiengesellschaft | A method for detecting a weather condition in an environment of a motor vehicle and control device and motor vehicle |
| CN108513674A (en)* | 2018-03-26 | 2018-09-07 | 深圳市锐明技术股份有限公司 | A kind of detection alarm method, storage medium and the server of Chinese herbaceous peony accumulated snow and icing |
| CN108701396A (en)* | 2018-03-26 | 2018-10-23 | 深圳市锐明技术股份有限公司 | A kind of detection alarm method, storage medium and the server of Chinese herbaceous peony accumulated snow and icing |
| CN108921013A (en)* | 2018-05-16 | 2018-11-30 | 浙江零跑科技有限公司 | A kind of visual scene identifying system and method based on deep neural network |
| CN108875593A (en)* | 2018-05-28 | 2018-11-23 | 上海交通大学 | Visible images weather recognition methods based on convolutional neural networks |
| CN108805198A (en)* | 2018-06-08 | 2018-11-13 | Oppo广东移动通信有限公司 | Image processing method, device, computer readable storage medium and electronic equipment |
| CN109447018A (en)* | 2018-11-08 | 2019-03-08 | 天津理工大学 | A kind of road environment visual perception method based on improvement Faster R-CNN |
| CN109525901A (en)* | 2018-11-27 | 2019-03-26 | Oppo广东移动通信有限公司 | Method for processing video frequency, device, electronic equipment and computer-readable medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807549A (en)* | 2019-10-29 | 2020-02-18 | 国网电力科学研究院武汉南瑞有限责任公司 | Generation method, generation device, generation system and electronic equipment of meteorological prediction model |
| CN111178407A (en)* | 2019-12-19 | 2020-05-19 | 中国平安人寿保险股份有限公司 | Road condition data screening method and device, computer equipment and storage medium |
| CN111178407B (en)* | 2019-12-19 | 2024-09-06 | 中国平安人寿保险股份有限公司 | Road condition data screening method, device, computer equipment and storage medium |
| CN113887286A (en)* | 2021-08-31 | 2022-01-04 | 际络科技(上海)有限公司 | Driver behavior monitoring method based on online video understanding network |
| CN115148040A (en)* | 2022-06-28 | 2022-10-04 | 东莞中科云计算研究院 | Unmanned vehicle control method and system for closed road environment |
| Publication number | Publication date |
|---|---|
| WO2020224117A1 (en) | 2020-11-12 |
| Publication | Publication Date | Title |
|---|---|---|
| CN112417953B (en) | Road condition detection and map data updating method, device, system and equipment | |
| CN113468967B (en) | Attention mechanism-based lane line detection method, attention mechanism-based lane line detection device, attention mechanism-based lane line detection equipment and attention mechanism-based lane line detection medium | |
| CN110186471A (en) | Air navigation aid, device, computer equipment and storage medium based on history video | |
| CN110471058A (en) | The system and method detected automatically for trailer attribute | |
| WO2022166625A1 (en) | Method for information pushing in vehicle travel scenario, and related apparatus | |
| CN111209780A (en) | Lane line attribute detection method, device, electronic device and readable storage medium | |
| WO2020042658A1 (en) | Data processing method, device, apparatus, and system | |
| US12056815B2 (en) | Generating map features based on aerial data and telemetry data | |
| CN112037142B (en) | Image denoising method, device, computer and readable storage medium | |
| CN114385662B (en) | Road network updating method and device, storage medium and electronic equipment | |
| US20210303886A1 (en) | Semantically-consistent augmented training data for traffic light detection | |
| CN110782653A (en) | Road information acquisition method and system | |
| CN113486856A (en) | Driver irregular behavior detection method based on semantic segmentation and convolutional neural network | |
| US11816543B2 (en) | System and method for using knowledge gathered by a vehicle | |
| CN117351712A (en) | Zhou Che track prediction method and system based on Cro-IntntFormer and fusing vehicle driving intention | |
| CN115631108A (en) | RGBD-based image defogging method and related equipment | |
| WO2024092590A1 (en) | Image processing method and apparatus, model training method and apparatus, and terminal device | |
| CN112163544A (en) | Method and system for judging random placement of non-motor vehicles | |
| CN114120645B (en) | A traffic scene extraction method in natural driving environment | |
| CN114419018A (en) | Image sampling method, system, device and medium | |
| CN111160282B (en) | Traffic light detection method based on binary Yolov3 network | |
| CN116612385B (en) | Remote sensing image multiclass information extraction method and system based on depth high-resolution relation graph convolution | |
| CN116434173B (en) | Road image detection method, device, electronic device and storage medium | |
| CN117831000A (en) | Traffic light detection method and device, electronic equipment and storage medium | |
| CN117078997A (en) | Image processing or training method, device, equipment and medium of image processing model |
| 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 | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20190830 | |
| RJ01 | Rejection of invention patent application after publication |