技术领域Technical Field
本申请涉及监控系统技术领域，特别的涉及一种基于智能分析设备与业务模型的告警识别方法及装置。The present application relates to the technical field of monitoring systems, and in particular to an alarm identification method and device based on intelligent analysis equipment and business models.
背景技术Background Art
针对各个行业对重点区域防控的诉求，如智慧城市、智慧军营、机场监控等场景，需要通过多种感知探测设备对管控区域实现全天候监测以及精准识别预警。然而，多源设备上报的告警容易受设备自身以及环境等多方因素影响，进而导致误报率居高不下，例如，设备部分元器件损害、风吹草动以及动物误碰等情况均会导致误报的产生。与此同时，从业务应用角度考虑，应尽量减少漏报，那么大量产生误报就成为不可避免的问题。此外，由于现有设备产生的原始报警类型固化单一，但同一类告警基于不同布防位置需要识别的情况可能不一样，因而不能满足用户基于时间、地点、目标等差异关注不同的业务事件的需求。例如，智能相机可以识别穿越警戒线，但实际用户需要识别船舶靠岸、船舶离岸以及管控区域人员越界等更具体更具有差异性的情况。In response to the demands of various industries for key area control, such as smart cities, smart military camps, airport monitoring and other scenarios, it is necessary to use a variety of sensing and detection equipment to achieve all-weather monitoring and accurate identification and early warning of the control area. However, the alarms reported by multi-source devices are easily affected by multiple factors such as the equipment itself and the environment, which leads to a high false alarm rate. For example, damage to some components of the equipment, the slightest disturbance, and accidental contact with animals can all lead to false alarms. At the same time, from the perspective of business applications, missed reports should be minimized, so a large number of false alarms become an inevitable problem. In addition, since the original alarm types generated by existing equipment are solidified and single, but the same type of alarm may need to be identified based on different deployment positions. The situation may be different, so it cannot meet the user's needs to pay attention to different business events based on differences in time, location, and target. For example, smart cameras can identify crossing the cordon, but actual users need to identify more specific and different situations such as ships docking, ships leaving the shore, and people crossing the boundary in the control area.
现有的报警系统通常是通过人工查看报警所在区域的视频画面进行核实分析是否误报，以降低误报率。不仅大大增加了用户的工作量，还因经常性的误报严重影响了用户的感知，使得用户丧失了对危险的紧迫感，进而可能因麻痹大意酿成大错。Existing alarm systems usually rely on manual verification of video footage of the alarm area to determine if it is a false alarm, in order to reduce the false alarm rate. This not only greatly increases the workload of users, but also seriously affects user perception due to frequent false alarms, causing users to lose their sense of urgency about danger, which may lead to serious mistakes due to carelessness.
与此同时，现有的报警系统往往直接将设备的原始告警呈现给用户或只通过多源设备的简单协同将告警传输给用户，没有基于业务所需进行深度的业务模型转换，因而不能满足用户差异化的识别防控的需求。At the same time, existing alarm systems often directly present the original alarms of the equipment to users or only transmit the alarms to users through simple collaboration of multi-source devices. There is no in-depth business model conversion based on business needs, and therefore cannot meet users' differentiated identification and prevention needs.
发明内容Summary of the invention
为解决上述提到的增加用户的工作量、严重影响用户的感知以及不能满足用户差异化的识别防控的需求的问题，本申请实施例提供了一种基于智能分析设备与业务模型的告警识别方法及装置，其技术方案如下：In order to solve the above-mentioned problems of increasing the workload of users, seriously affecting the perception of users, and failing to meet the differentiated identification and prevention needs of users, the embodiment of the present application provides an alarm identification method and device based on intelligent analysis equipment and business models, and its technical solution is as follows:
第一方面，本申请实施例提供了一种基于智能分析设备与业务模型的告警识别方法，包括：In a first aspect, an embodiment of the present application provides an alarm identification method based on an intelligent analysis device and a business model, including:
当接收到由智能分析设备发送的至少一个告警图像时，对告警图像进行信息提取处理，得到至少两种信息；When receiving at least one warning image sent by the intelligent analysis device, performing information extraction processing on the warning image to obtain at least two types of information;
基于至少两种信息以及预设的业务模型，得到事件描述；其中，预设的业务模型由至少两种样本信息以及与样本信息对应的事件描述训练得到；Based on at least two kinds of information and a preset business model, an event description is obtained; wherein the preset business model is trained by at least two kinds of sample information and event descriptions corresponding to the sample information;
基于至少两种信息以及预设的等级划分表，得到事件等级，并按照与事件等级对应的发送方式，将事件描述发送至用户客户端；其中，等级划分表至少包括两种信息以及与信息对应的等级。Based on at least two types of information and a preset level classification table, an event level is obtained, and the event description is sent to the user client in a sending mode corresponding to the event level; wherein the level classification table includes at least two types of information and levels corresponding to the information.
在第一方面的一种可选方案中，在接收到由智能分析设备发送的至少一个告警图像之前，还包括：In an optional solution of the first aspect, before receiving at least one warning image sent by the intelligent analysis device, the method further includes:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型属于预设的类型范围时，将初始图像发送给智能分析设备，以由智能分析设备对初始图像进行筛选处理并得到至少一个告警图像；其中，预设的类型范围至少包括两种感知探测设备的类型。When it is identified that the type of the perception detection device belongs to a preset type range, the initial image is sent to the intelligent analysis device, so that the intelligent analysis device screens and processes the initial image and obtains at least one alarm image; wherein the preset type range includes at least two types of perception detection devices.
在第一方面的又一种可选方案中，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：In yet another optional solution of the first aspect, before obtaining the event description based on the at least two types of information and the preset business model, the method further includes:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型不属于预设的类型范围时，对初始图像进行信息提取处理，得到至少两种信息。When it is identified that the type of the perception detection device does not belong to the preset type range, information extraction processing is performed on the initial image to obtain at least two types of information.
在第一方面的又一种可选方案中，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：In yet another optional solution of the first aspect, before obtaining the event description based on the at least two types of information and the preset business model, the method further includes:
当接收到由智能分析设备发送的至少一个告警视频时，获取告警视频中任意两个相邻帧图像；When at least one alarm video sent by the intelligent analysis device is received, any two adjacent frame images in the alarm video are obtained;
计算任意两个相邻帧图像之间的像素差值；Calculate the pixel difference between any two adjacent frame images;
当检测到任意一个像素差值大于预设的第一阈值时，对像素差值所对应的后帧图像进行信息提取处理，得到至少两种信息；其中，后帧图像为任意两个相邻帧图像中在时间上后出现的一帧图像。When it is detected that any pixel difference is greater than a preset first threshold, information extraction processing is performed on the subsequent frame image corresponding to the pixel difference to obtain at least two types of information; wherein the subsequent frame image is a frame image that appears later in time between any two adjacent frame images.
在第一方面的又一种可选方案中，至少两种信息包括行为信息以及区域信息；In yet another optional solution of the first aspect, the at least two types of information include behavior information and area information;
基于至少两种信息以及预设的业务模型，得到事件描述，包括：Based on at least two types of information and a preset business model, an event description is obtained, including:
当检测到至少两种信息中存在行为信息时，获取区域信息的重要等级，并根据重要等级确定出描述顺序；When it is detected that there is behavior information in at least two types of information, obtaining the importance level of the area information, and determining the description order according to the importance level;
将描述顺序、行为信息以及区域信息输入至预设的业务模型，得到事件描述。The description order, behavior information, and region information are input into a preset business model to obtain an event description.
在第一方面的又一种可选方案中，基于至少两种信息以及预设的等级划分表，得到事件等级，包括：In another optional solution of the first aspect, obtaining the event level based on at least two types of information and a preset level classification table includes:
基于预设的等级划分表，得到与每种信息对应的信息等级；Based on a preset level classification table, obtain the information level corresponding to each type of information;
当检测到信息等级与预设的第一等级相同的信息的种类数量大于或等于预设的第二阈值时，将一级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is greater than or equal to a preset second threshold, the first level is taken as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量大于或等于预设的第二阈值时，将二级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is greater than or equal to the preset second threshold, the second level is used as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量小于预设的第二阈值时，将三级作为事件等级。When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is less than the preset second threshold, level three is used as the event level.
在第一方面的又一种可选方案中，按照与事件等级对应的发送方式，将事件描述发送至用户客户端，包括：In another optional solution of the first aspect, sending the event description to the user client in a sending manner corresponding to the event level includes:
当识别出事件等级为一级时，按照预设的第一时间间隔以及电话呼叫方式，将事件描述发送至用户客户端；When the event level is identified as level one, the event description is sent to the user client according to a preset first time interval and telephone call method;
当识别出事件等级为二级时，按照预设的第二时间间隔以及语音助手播报方式，将事件描述发送至用户客户端；其中，第二时间间隔大于第一时间间隔；When the event level is identified as level 2, the event description is sent to the user client according to the preset second time interval and voice assistant broadcast mode; wherein the second time interval is greater than the first time interval;
当识别出事件等级为三级时，按照预设的第三时间间隔以及电子邮件通知方式，将事件描述发送至用户客户端；其中，第三时间间隔大于第二时间间隔。When the event level is identified as level three, the event description is sent to the user client according to a preset third time interval and email notification method; wherein the third time interval is greater than the second time interval.
第二方面，本申请实施例提供了一种基于智能分析设备与业务模型的告警识别装置，包括：In a second aspect, an embodiment of the present application provides an alarm identification device based on an intelligent analysis device and a business model, including:
第一处理模块，用于当接收到由智能分析设备发送的至少一个告警图像时，对告警图像进行信息提取处理，得到至少两种信息；A first processing module is used to extract information from the warning image when receiving at least one warning image sent by the intelligent analysis device to obtain at least two types of information;
第二处理模块，用于基于至少两种信息以及预设的业务模型，得到事件描述；其中，预设的业务模型由至少两种样本信息以及与样本信息对应的事件描述训练得到；A second processing module is used to obtain an event description based on at least two types of information and a preset business model; wherein the preset business model is trained by at least two types of sample information and event descriptions corresponding to the sample information;
第三处理模块，用于基于至少两种信息以及预设的等级划分表，得到事件等级，并按照与事件等级对应的发送方式，将事件描述发送至用户客户端；其中，等级划分表至少包括两种信息以及与信息对应的等级。The third processing module is used to obtain the event level based on at least two types of information and a preset level classification table, and send the event description to the user client in a sending method corresponding to the event level; wherein the level classification table includes at least two types of information and the levels corresponding to the information.
第三方面，本申请实施例还提供了一种基于智能分析设备与业务模型的告警识别装置，包括处理器以及存储器；In a third aspect, an embodiment of the present application further provides an alarm identification device based on an intelligent analysis device and a business model, including a processor and a memory;
处理器与存储器连接；The processor is connected to the memory;
存储器，用于存储可执行程序代码；A memory for storing executable program codes;
处理器通过读取存储器中存储的可执行程序代码来运行与可执行程序代码对应的程序，以用于实现本申请实施例第一方面或第一方面的任意一种实现方式提供的基于智能分析设备与业务模型的告警识别方法。The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the alarm identification method based on intelligent analysis equipment and business model provided by the first aspect of the embodiment of the present application or any implementation method of the first aspect.
第四方面，本申请实施例提供了一种计算机存储介质，计算机存储介质存储有计算机程序，计算机程序包括程序指令，程序指令当被处理器执行时，可实现本申请实施例第一方面或第一方面的任意一种实现方式提供的基于智能分析设备与业务模型的告警识别方法。In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the alarm identification method based on intelligent analysis equipment and business models provided by the first aspect of the embodiment of the present application or any implementation method of the first aspect can be implemented.
本说明书一些实施例提供的技术方案带来的有益效果至少包括：The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least:
在基于智能分析设备与业务模型的告警识别过程中，通过接收由智能分析设备复核分析后发送的告警图像而不是所有被采集到的图像，消除了采集设备自身以及环境等多方因素的影响，从而降低了误报率并改善了用户的感知，与此同时，还根据该图像包含的信息以及预设的业务模型确定出该图像中所发生的事件的描述内容（即告警内容）以及该事件的重要程度，进而根据该事件的重要程度，采用合适的发送方式将该事件的描述内容发送至用户客户端，提供给了用户更精准的告警内容，满足了用户差异化的识别防控的需求。In the alarm identification process based on intelligent analysis equipment and business models, by receiving the alarm image sent after review and analysis by the intelligent analysis equipment instead of all the collected images, the influence of multiple factors such as the collection equipment itself and the environment is eliminated, thereby reducing the false alarm rate and improving the user's perception. At the same time, the description content of the event occurring in the image (i.e., the alarm content) and the importance of the event are determined based on the information contained in the image and the preset business model. Then, according to the importance of the event, the description content of the event is sent to the user client in an appropriate sending manner, providing the user with more accurate alarm content and meeting the user's differentiated identification and prevention needs.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，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 more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
图1为本申请实施例提供的一种基于智能分析设备与业务模型的告警识别方法的整体流程图；FIG1 is an overall flow chart of an alarm identification method based on intelligent analysis equipment and business models provided in an embodiment of the present application;
图2为本申请实施例提供的一种业务模型的训练样本的获取方法的整体流程图；FIG2 is an overall flow chart of a method for acquiring training samples of a business model provided in an embodiment of the present application;
图3为本申请实施例提供的又一种基于智能分析设备与业务模型的告警识别方法的整体流程图；FIG3 is an overall flow chart of another alarm identification method based on intelligent analysis equipment and business models provided in an embodiment of the present application;
图4为本申请实施例提供的一种基于智能分析设备与业务模型的告警识别装置的结构示意图；FIG4 is a schematic diagram of the structure of an alarm identification device based on an intelligent analysis device and a business model provided in an embodiment of the present application;
图5为本申请实施例提供的又一种基于智能分析设备与业务模型的告警识别装置的结构示意图。FIG5 is a schematic diagram of the structure of another alarm identification device based on intelligent analysis equipment and business models provided in an embodiment of the present application.
具体实施方式DETAILED DESCRIPTION
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述。The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
在下述介绍中，术语“第一”、“第二”仅为用于描述的目的，而不能理解为指示或暗示相对重要性。下述介绍提供了本申请的多个实施例，不同实施例之间可以替换或者合并组合，因此本申请也可认为包含所记载的相同和/或不同实施例的所有可能组合。因而，如果一个实施例包含特征A、B、C，另一个实施例包含特征B、D，那么本申请也应视为包括含有A、B、C、D的一个或多个所有其他可能的组合的实施例，尽管该实施例可能并未在以下内容中有明确的文字记载。In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and/or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, even though the embodiment may not be clearly recorded in the following text.
下面的描述提供了示例，并且不对权利要求书中阐述的范围、适用性或示例进行限制。可以在不脱离本申请内容的范围的情况下，对描述的元素的功能和布置做出改变。各个示例可以适当省略、替代或添加各种过程或组件。例如所描述的方法可以以所描述的顺序不同的顺序来执行，并且可以添加、省略或组合各种步骤。此外，可以将关于一些示例描述的特征组合到其他示例中。The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.
请参阅图1，图1示出了本申请实施例提供的一种基于智能分析设备与业务模型的告警识别方法的整体流程图。Please refer to FIG. 1 , which shows an overall flow chart of an alarm identification method based on intelligent analysis equipment and business models provided in an embodiment of the present application.
如图1所示，该基于智能分析设备与业务模型的告警识别方法至少可以包括以下步骤：As shown in FIG1 , the alarm identification method based on intelligent analysis equipment and business model may at least include the following steps:
步骤101、当接收到由智能分析设备发送的至少一个告警图像时，对告警图像进行信息提取处理，得到至少两种信息。Step 101: When at least one warning image sent by an intelligent analysis device is received, information extraction processing is performed on the warning image to obtain at least two types of information.
在本申请实施例中，基于智能分析设备与业务模型的告警识别方法可以但不局限于应用在服务器以及云平台，其中，智能分析设备是指运用计算机视觉、机器学习、大数据分析等技术，对原始数据（尤其是图像、视频、音频信号等非结构化数据）进行智能处理和分析的硬件设备，如智能安防摄像头、智能物联网传感器、智能视频分析服务器、边缘计算盒子、无人机智能分析系统、智能交通管理系统以及智能医疗影像分析设备等；业务模型是指根据业务需求自定义的一个经过训练的模型，可由至少两种样本信息以及与样本信息对应的事件描述训练得到，该模型的输入数据可以但不局限于为文本内容或数值型数据，输出数据可以但不局限于为文本内容。In an embodiment of the present application, the alarm identification method based on intelligent analysis equipment and business models can be but is not limited to application in servers and cloud platforms, wherein the intelligent analysis equipment refers to a hardware device that uses computer vision, machine learning, big data analysis and other technologies to intelligently process and analyze raw data (especially unstructured data such as images, videos, audio signals, etc.), such as smart security cameras, smart Internet of Things sensors, smart video analysis servers, edge computing boxes, drone intelligent analysis systems, smart traffic management systems, and smart medical imaging analysis equipment, etc.; the business model refers to a trained model customized according to business needs, which can be trained by at least two types of sample information and event descriptions corresponding to the sample information. The input data of the model can be but is not limited to text content or numerical data, and the output data can be but is not limited to text content.
在基于智能分析设备与业务模型的告警识别过程中，通过接收由智能分析设备复核分析后发送的告警图像而不是所有被采集到的图像，消除了采集设备自身以及环境等多方因素的影响，从而降低了误报率并改善了用户的感知，与此同时，还根据该图像包含的信息以及预设的业务模型确定出该图像中所发生的事件的描述内容（即告警内容）以及该事件的重要程度，进而根据该事件的重要程度，采用合适的发送方式将该事件的描述内容发送至用户客户端，提供给了用户更精准的告警内容，满足了用户差异化的识别防控的需求。In the alarm identification process based on intelligent analysis equipment and business models, by receiving the alarm image sent after review and analysis by the intelligent analysis equipment instead of all the collected images, the influence of multiple factors such as the collection equipment itself and the environment is eliminated, thereby reducing the false alarm rate and improving the user's perception. At the same time, the description content of the event occurring in the image (i.e., the alarm content) and the importance of the event are determined based on the information contained in the image and the preset business model. Then, according to the importance of the event, the description content of the event is sent to the user client in an appropriate sending manner, providing the user with more accurate alarm content and meeting the user's differentiated identification and prevention needs.
具体地，在基于智能分析设备与业务模型的告警识别过程中，当接收到由智能分析设备进行复核分析后发送的至少一个告警图像时，可对该告警图像进行信息提取处理，进而得到至少两种信息。其中，信息提取处理的方式可以但不局限于为物体检测算法（如Faster R-CNN、YOLOv3、SSD等）以及深度神经网络模型（如卷积神经网络CNN、循环神经网络RNN等）。Specifically, in the alarm identification process based on the intelligent analysis device and the business model, when at least one alarm image sent by the intelligent analysis device after review and analysis is received, the alarm image can be subjected to information extraction processing to obtain at least two types of information. The information extraction processing method can be, but is not limited to, object detection algorithms (such as Faster R-CNN, YOLOv3, SSD, etc.) and deep neural network models (such as convolutional neural network CNN, recurrent neural network RNN, etc.).
可以理解的是，得到的至少两种信息可以但不局限于为对象信息（如人、车、船、无人机、牲畜等）、区域信息（如营区门口、监控室、管理处、中心办公区等）、行为信息（如下蹲、攀爬、人员聚集、打架斗殴等）以及配置信息（如持枪、穿军装、是否骑行、是否拎包、是否带帽、性别、国别等）。It is understandable that the at least two types of information obtained can be but are not limited to object information (such as people, vehicles, ships, drones, livestock, etc.), area information (such as camp gate, monitoring room, management office, central office area, etc.), behavior information (such as squatting, climbing, gathering of people, fighting, etc.) and configuration information (such as holding a gun, wearing a military uniform, riding a bicycle, carrying a bag, wearing a hat, gender, country, etc.).
作为本申请实施例的一种可选，在接收到由智能分析设备发送的至少一个告警图像之前，还包括：As an option of the embodiment of the present application, before receiving at least one warning image sent by the intelligent analysis device, the method further includes:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型属于预设的类型范围时，将初始图像发送给智能分析设备，以由智能分析设备对初始图像进行筛选处理并得到至少一个告警图像；其中，预设的类型范围至少包括两种感知探测设备的类型。When it is identified that the type of the perception detection device belongs to a preset type range, the initial image is sent to the intelligent analysis device, so that the intelligent analysis device screens and processes the initial image and obtains at least one alarm image; wherein the preset type range includes at least two types of perception detection devices.
具体地，在接受到由智能分析设备发送的告警图像之前，当接收到由感知探测设备发送的初始图像时，可对该感知探测设备的类型进行判断。若该感知探测设备属于预设的类型范围（如简易雷达、分辨率较低的摄像头、震动传感器、金属探测器、红外探测器等），代表该感知探测设备对目标（即上述对象信息）的识别度不够准确，将导致获取到的初始图像不一定存在目标（即误报情况），例如，当阳光、照明变动或反光物体等照射到红外探测器上时，将导致与该红外探测器协同作用的成像设备进行拍摄，进而获取到不存在目标的初始图像。因此，此时可将该初始图像发送给智能分析设备，并由智能分析设备通过目标识别算法对其进行复核分析，以确认该初始图像中是否存在目标，进而将存在目标的初始图像作为告警图像。Specifically, before receiving the warning image sent by the intelligent analysis device, when receiving the initial image sent by the perception detection device, the type of the perception detection device can be judged. If the perception detection device belongs to the preset type range (such as a simple radar, a camera with low resolution, a vibration sensor, a metal detector, an infrared detector, etc.), it means that the recognition degree of the target (i.e., the above-mentioned object information) of the perception detection device is not accurate enough, which will cause the initial image obtained to not necessarily contain the target (i.e., a false alarm situation). For example, when sunlight, lighting changes, or reflective objects, etc. shine on the infrared detector, it will cause the imaging device that cooperates with the infrared detector to take pictures, and then obtain the initial image without the target. Therefore, at this time, the initial image can be sent to the intelligent analysis device, and the intelligent analysis device will review and analyze it through the target recognition algorithm to confirm whether there is a target in the initial image, and then use the initial image with the target as the warning image.
需要说明的是，感知探测设备可以但不局限于为一个设备（如监控摄像头）或由多个设备协同组成（如定位型振动光纤联动附近摄像机）；目标识别算法可以但不局限于为模板匹配算法、关键点检测与描述算法、实例分割算法以及度量学习算法等；预设的类型范围至少包括两种感知探测设备的类型。It should be noted that the perception and detection device can be but is not limited to being a single device (such as a surveillance camera) or being composed of multiple devices working together (such as a positioning vibration optical fiber linked to nearby cameras); the target recognition algorithm can be but is not limited to being a template matching algorithm, a key point detection and description algorithm, an instance segmentation algorithm, and a metric learning algorithm, etc.; the preset type range includes at least two types of perception and detection devices.
作为本申请实施例的一种可选，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：As an option of the embodiment of the present application, before obtaining the event description based on at least two kinds of information and a preset business model, the method further includes:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型不属于预设的类型范围时，对初始图像进行信息提取处理，得到至少两种信息。When it is identified that the type of the perception detection device does not belong to the preset type range, information extraction processing is performed on the initial image to obtain at least two types of information.
具体地，在根据至少两种信息以及预设的业务模型得到相应的事件描述之前，当接收到由感知探测设备发送的初始图像时，可对该感知探测设备的类型进行判断。若该感知探测设备不属于预设的类型范围，代表该感知探测设备对目标的识别度较为准确，获取到的初始图像是存在目标的，因而可直接对该初始图像进行信息提取处理，从而得到至少两种信息。Specifically, before obtaining the corresponding event description according to at least two kinds of information and a preset business model, when receiving the initial image sent by the perception detection device, the type of the perception detection device can be judged. If the perception detection device does not belong to the preset type range, it means that the perception detection device has a relatively accurate recognition degree of the target, and the acquired initial image has the target, so the initial image can be directly subjected to information extraction processing to obtain at least two kinds of information.
需要说明的是，除可根据感知探测设备的类型对初始图像进行初步筛选之外，还可根据具体的应用场景或业务需求选择对应的初始图像，进而发送给智能分析设备，并由智能分析设备对其进行复核分析，从而得到存在目标的告警图像。It should be noted that in addition to preliminary screening of the initial image according to the type of perception detection equipment, the corresponding initial image can also be selected according to the specific application scenario or business needs, and then sent to the intelligent analysis device, which will review and analyze it to obtain an alarm image of the target.
作为本申请实施例的一种可选，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：As an option of the embodiment of the present application, before obtaining the event description based on at least two kinds of information and a preset business model, the method further includes:
当接收到由智能分析设备发送的至少一个告警视频时，获取告警视频中任意两个相邻帧图像；When at least one alarm video sent by the intelligent analysis device is received, any two adjacent frame images in the alarm video are obtained;
计算任意两个相邻帧图像之间的像素差值；Calculate the pixel difference between any two adjacent frame images;
当检测到任意一个像素差值大于预设的第一阈值时，对像素差值所对应的后帧图像进行信息提取处理，得到至少两种信息；其中，后帧图像为任意两个相邻帧图像中在时间上后出现的一帧图像。When it is detected that any pixel difference is greater than a preset first threshold, information extraction processing is performed on the subsequent frame image corresponding to the pixel difference to obtain at least two types of information; wherein the subsequent frame image is a frame image that appears later in time between any two adjacent frame images.
具体地，在根据至少两种信息以及预设的业务模型得到相应的事件描述之前，当接收到由智能分析设备发送的至少一个告警视频时，可以但不局限于通过视频处理软件（如Adobe Premiere Pro、Final CutPro、VLC Media Player等）逐帧导出图像。Specifically, before obtaining the corresponding event description based on at least two types of information and a preset business model, when at least one alarm video sent by the intelligent analysis device is received, the image can be exported frame by frame through but not limited to video processing software (such as Adobe Premiere Pro, Final CutPro, VLC Media Player, etc.).
接着，可以但不局限于通过OpenCV软件库中的absdiff()函数计算任意两个相邻帧图像之间的像素差值，当检测到任意一个像素差值大于预设的第一阈值时，代表图像内容发生了显著变化，也意味着可能存在入侵情况，此时，可对该像素差值所对应的后帧图像进行信息提取处理，进而得到至少两种信息。其中，预设的第一阈值可根据具体情况进行设定，后帧图像为任意两个相邻帧图像中时间上后出现的一帧图像。Next, the pixel difference between any two adjacent frame images can be calculated by the absdiff() function in the OpenCV software library, but is not limited to it. When any pixel difference is detected to be greater than a preset first threshold, it means that the image content has changed significantly, which also means that there may be an intrusion. At this time, the subsequent frame image corresponding to the pixel difference can be processed for information extraction, thereby obtaining at least two types of information. Among them, the preset first threshold can be set according to the specific situation, and the subsequent frame image is a frame image that appears later in time between any two adjacent frame images.
需要说明的是，如果使用灰度图像进行计算，阈值可设置在5~20之间；如使用彩色图像进行计算，阈值可设置在10~50之间，具体数值可根据多次样本实验来确定。It should be noted that if a grayscale image is used for calculation, the threshold can be set between 5 and 20; if a color image is used for calculation, the threshold can be set between 10 and 50. The specific value can be determined based on multiple sample experiments.
值得注意的是，上述过程旨在获取到告警视频中的关键帧，关键帧往往对应于视频序列中图像内容发生显著变化的时刻（如出现入侵情况的时刻）。该时刻可以但不局限于通过计算相邻帧间的像素差值、结构相似性指数（SSIM）、峰值信噪比（PSNR）等图像相似度指标来确定。It is worth noting that the above process aims to obtain the key frames in the alarm video, which often correspond to the moment when the image content in the video sequence changes significantly (such as the moment when an intrusion occurs). This moment can be determined by calculating, but is not limited to, image similarity indicators such as pixel difference between adjacent frames, structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR).
步骤102、基于至少两种信息以及预设的业务模型，得到事件描述。Step 102: Get an event description based on at least two types of information and a preset business model.
具体地，在得到至少两种信息之后，由于该信息不一定是数值型数据，业务模型对于数值型的数据的处理效果更为准确，因而可将该信息转换为对应的数值型数据，并将该数值型数据输入至预设的业务模型，从而生成与事件描述对应的数值型数据，并对其进行转换处理，得到相应的文字内容（即事件描述）。其中，预设的业务模型由至少两种样本信息以及与样本信息对应的事件描述训练得到，事件描述可以但不局限于为一个短语、一个语句或多个语句，如“营区门口发现可疑人员”、“1号码头有船舶靠岸”、“中心办公区发现人员入侵”。Specifically, after obtaining at least two types of information, since the information is not necessarily numerical data, the business model has a more accurate processing effect on numerical data, so the information can be converted into corresponding numerical data, and the numerical data is input into the preset business model, thereby generating numerical data corresponding to the event description, and converting it to obtain the corresponding text content (i.e., event description). Among them, the preset business model is trained by at least two types of sample information and event descriptions corresponding to the sample information. The event description can be, but is not limited to, a phrase, a sentence, or multiple sentences, such as "suspicious persons are found at the entrance of the camp", "ships are docked at Pier 1", and "personnel intrusion is found in the central office area".
可以理解的是，将非数值型数据（如类别标签、文本、日期等）转化为数值形式可以但不局限于采用独热编码、标签编码、词嵌入、时间序列编码等方法。It is understandable that converting non-numeric data (such as category labels, text, dates, etc.) into numerical form can be, but is not limited to, using methods such as one-hot encoding, label encoding, word embedding, time series encoding, etc.
需要说明的是，对于不同的场景，感知探测设备的类型可能不同，其对应的时间描述也不同。例如，对于保险柜、墙壁等易受到敲击、震动或强行撬动的情况，可由震动传感器与监控摄像头协同组成感知探测设备，其对应的告警图像中的目标类型主要为人，相应的事件描述内容主要为某区域发现人员入侵；对于水中或空中的目标检测，可采用红外热像仪作为感知探测设备，其对应的告警图像中的目标类型主要为船只、飞行器等，相应的事件描述内容主要为某码头发现船舶靠岸、某区域上空发现不明飞行物等。It should be noted that for different scenarios, the type of perception detection equipment may be different, and the corresponding time description is also different. For example, for safes, walls, etc. that are susceptible to knocking, vibration, or forced prying, the perception detection equipment can be composed of a vibration sensor and a surveillance camera. The target type in the corresponding alarm image is mainly people, and the corresponding event description content is mainly that a person is found to be intruding in a certain area; for target detection in water or in the air, an infrared thermal imager can be used as a perception detection device. The target type in the corresponding alarm image is mainly ships, aircraft, etc., and the corresponding event description content is mainly that a ship is found docking at a certain dock, and an unidentified flying object is found over a certain area.
此处，可参阅图2，图2示出了本申请实施例提供的一种业务模型的训练样本的获取方法的整体流程图。Here, please refer to Figure 2, which shows an overall flow chart of a method for acquiring training samples of a business model provided in an embodiment of the present application.
如图2所示，训练样本可以包括“事件名称”、“事件等级”、“提醒方式”、“事件来源”“是否需要二次确认”以及“转换条件”等内容。As shown in FIG. 2 , the training samples may include “event name”, “event level”, “reminder method”, “event source”, “whether secondary confirmation is required”, and “conversion condition”.
需要说明的是，事件名称即上述事件描述，主要用于描述拍摄到入侵事件的图像或视频的内容；事件等级用于表明该入侵事件的严重性以及确定出对应的通知用户的方式（即发送事件描述的方式）；提醒方式主要用于保证用户能够及时关注到入侵事件的发生；事件来源主要用于确定出事件发生的位置；是否需要二次确认主要用于提高告警图像或告警视频中存在目标的识别效率；转换条件主要用于确定出对应的事件名称。It should be noted that the event name, namely the above-mentioned event description, is mainly used to describe the content of the image or video captured of the intrusion event; the event level is used to indicate the severity of the intrusion event and to determine the corresponding method of notifying the user (i.e. the method of sending the event description); the reminder method is mainly used to ensure that the user can pay attention to the occurrence of the intrusion event in time; the event source is mainly used to determine the location where the event occurred; whether secondary confirmation is required is mainly used to improve the recognition efficiency of targets in the alarm image or alarm video; the conversion condition is mainly used to determine the corresponding event name.
其中，事件名称可由用户根据具体环境（如具体的区域划分，具体的地貌特征等）确定出；事件等级可根据具体的目标类型（如人、车、船等）确定出；提醒方式可根据事件等级确定出（如事件等级较高时播放音频能更显著地提醒用户该入侵事件的发生）；事件来源可根据接收到告警图像或告警视频的设备确定出；是否需要二次确认可由中心平台根据事件来源确定出；转换条件可由用户根据事件名称所包含的关键信息确定出。Among them, the event name can be determined by the user according to the specific environment (such as specific area division, specific landform features, etc.); the event level can be determined according to the specific target type (such as people, cars, ships, etc.); the reminder method can be determined according to the event level (such as playing audio when the event level is high can more significantly remind the user of the occurrence of the intrusion event); the event source can be determined according to the device that receives the alarm image or alarm video; whether secondary confirmation is required can be determined by the central platform according to the source of the event; the conversion condition can be determined by the user according to the key information contained in the event name.
可以理解的是，上述训练样本内容的获取过程可由图2中的五个步骤得到，即：自定义事件名称、事件等级；设置客户端是否弹窗提醒、是否播放音频；设置报警源；设置事件处理流程；设置转换条件。It can be understood that the process of obtaining the above training sample content can be obtained by the five steps in Figure 2, namely: customizing the event name and event level; setting whether the client pops up a reminder and whether to play audio; setting the alarm source; setting the event processing flow; and setting the conversion conditions.
此处，可参阅下述两个具体的训练样本：Here, you can refer to the following two specific training samples:
（1）事件名称：重点区域发现人员入侵；事件等级：高；是否弹窗提醒：是；是否播放音频：是；报警源：A区1号摄像机以及A区2号摄像机；事件处理流程：需要经过智能分析设备进行复核分析；转换条件：发生该入侵情况的区域为A区且识别出的目标类型为人。(1) Event name: Human intrusion detected in key areas; Event level: High; Pop-up window reminder: Yes; Play audio: Yes; Alarm source: Camera 1 in Area A and Camera 2 in Area A; Event processing flow: Needs to be reviewed and analyzed by intelligent analysis equipment; Conversion condition: The area where the intrusion occurred is Area A and the identified target type is human.
（2）事件名称：边界区域发现船舶靠岸；事件等级：中；是否弹窗提醒：是；是否播放音频：否；报警源：B区3号摄像机；事件处理流程：不需要经过智能分析设备进行复核分析；转换条件：发生该入侵情况的区域为B区且识别出的目标类型为船。(2) Event name: A ship was found docking in the border area; Event level: Medium; Pop-up reminder: Yes; Play audio: No; Alarm source: Camera No. 3 in Area B; Event processing flow: No need for review and analysis by intelligent analysis equipment; Conversion condition: The area where the intrusion occurred is Area B and the identified target type is a ship.
作为本申请实施例的一种可选，至少两种信息包括行为信息以及区域信息；As an option in the embodiment of the present application, the at least two types of information include behavior information and area information;
基于至少两种信息以及预设的业务模型，得到事件描述，包括：Based on at least two types of information and a preset business model, an event description is obtained, including:
当检测到至少两种信息中存在行为信息时，获取区域信息的重要等级，并根据重要等级确定出描述顺序；When it is detected that there is behavior information in at least two types of information, obtaining the importance level of the area information, and determining the description order according to the importance level;
将描述顺序、行为信息以及区域信息输入至预设的业务模型，得到事件描述。The description order, behavior information, and region information are input into a preset business model to obtain an event description.
具体地，在根据至少两种信息以及预设的业务模型得到相应的事件描述的过程中，可先对至少两种信息进行检测，当检测到至少两种信息中存在行为信息时，根据区域信息在预设的数据库中获取它的重要等级，其中，预设的数据库中存有至少两个不同区域的重要等级，例如，边界区域的重要等级为低，中心区域的重要等级为高或A区的重要等级为中，B区的重要等级为高等。Specifically, in the process of obtaining the corresponding event description based on at least two kinds of information and a preset business model, the at least two kinds of information can be detected first. When behavioral information is detected in at least two kinds of information, its importance level is obtained in a preset database based on the regional information, wherein the preset database stores the importance levels of at least two different areas, for example, the importance level of the boundary area is low, and the importance level of the central area is high, or the importance level of area A is medium, and the importance level of area B is high, etc.
可以理解的是，可按照与管控区域中心的距离将管控区域分为边界区域、内部区域以及中心区域或根据不同办公区或不同用途的场地将管控区域分为A区（如机要室）、B区（如会客区）等。It is understandable that the control area can be divided into a boundary area, an internal area and a central area according to the distance from the center of the control area, or can be divided into Area A (such as a confidential room), Area B (such as a meeting area), etc. according to different office areas or venues for different purposes.
接着，可根据该重要等级确定出描述顺序，其方式可以但不局限于为：当该区域信息的重要等级为高时，描述顺序为区域信息在行为信息之前，以提高用户对于区域信息的关注度；当该区域信息的重要等级为低时，描述顺序为区域信息在行为信息之后，以提高用户对于行为信息的关注度。Next, the description order may be determined based on the importance level, and the method may be, but is not limited to, as follows: when the importance level of the area information is high, the description order is that the area information is before the behavior information, so as to increase the user's attention to the area information; when the importance level of the area information is low, the description order is that the area information is after the behavior information, so as to increase the user's attention to the behavior information.
在确定出描述顺序之后，可先将描述顺序转换为更易识别的数值型数据，如用数字1代表描述顺序为区域信息在行为信息之前，用数字0代表描述顺序为区域信息在行为信息之后，进而可将与描述顺序对应的数值型数据、行为信息以及区域信息输入至预设的业务模型，从而得到遵循该描述顺序的文字内容（即事件描述）。After determining the description order, the description order can be converted into more easily recognizable numerical data, such as using the number 1 to represent the description order of area information before behavior information, and using the number 0 to represent the description order of area information after behavior information. Then, the numerical data, behavior information, and area information corresponding to the description order can be input into a preset business model to obtain text content (i.e., event description) that follows the description order.
步骤103、基于至少两种信息以及预设的等级划分表，得到事件等级，并按照与事件等级对应的发送方式，将事件描述发送至用户客户端。Step 103: Based on at least two types of information and a preset level classification table, obtain the event level, and send the event description to the user client in a sending manner corresponding to the event level.
具体地，在得到事件描述之后，可根据至少两种信息在预设的等级划分表中找出与信息对应的等级，其中，等级划分表至少包括两种信息以及与信息对应的等级。例如，对象信息为人时，对应的等级为高；对象信息为船时，对应的等级信息为中；区域信息为中心办公区时，对应的等级为高；区域信息为边界围栏时，对应的等级为低。Specifically, after obtaining the event description, the level corresponding to the information can be found in a preset level classification table according to at least two types of information, wherein the level classification table includes at least two types of information and the levels corresponding to the information. For example, when the object information is a person, the corresponding level is high; when the object information is a ship, the corresponding level information is medium; when the area information is a central office area, the corresponding level is high; when the area information is a boundary fence, the corresponding level is low.
进一步的，可根据所有信息所对应的等级判断出与该事件描述对应的事件等级，其方式可以但不局限于为：以所有信息所对应的等级中的最高等级作为事件等级。例如，当所有信息所对应的等级为低、中、中时，事件等级为中；当所有信息所对应的等级为低、中、高时，事件等级为高；当所有信息所对应的等级为低、低、低时，事件等级为低。Furthermore, the event level corresponding to the event description can be determined based on the levels corresponding to all the information, and the method can be, but is not limited to, taking the highest level among the levels corresponding to all the information as the event level. For example, when the levels corresponding to all the information are low, medium, and medium, the event level is medium; when the levels corresponding to all the information are low, medium, and high, the event level is high; when the levels corresponding to all the information are low, low, and low, the event level is low.
在确定出事件等级之后，可按照与该事件等级对应的发送方式，将事件描述发送到用户客户端，其中，发送方式可以但不局限于为电子邮件通知、短信通知、电话呼叫、移动应用推送通知、桌面弹窗、语音助手播报以及大屏展示等。After determining the event level, the event description can be sent to the user client in a sending method corresponding to the event level, where the sending method can be but is not limited to email notification, SMS notification, phone call, mobile application push notification, desktop pop-up window, voice assistant broadcast, and large-screen display.
作为本申请实施例的一种可选，基于至少两种信息以及预设的等级划分表，得到事件等级，包括：As an option of the embodiment of the present application, the event level is obtained based on at least two types of information and a preset level classification table, including:
基于预设的等级划分表，得到与每种信息对应的信息等级；Based on a preset level classification table, obtain the information level corresponding to each type of information;
当检测到信息等级与预设的第一等级相同的信息的种类数量大于或等于预设的第二阈值时，将一级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is greater than or equal to a preset second threshold, the first level is taken as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量大于或等于预设的第二阈值时，将二级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is greater than or equal to the preset second threshold, the second level is used as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量小于预设的第二阈值时，将三级作为事件等级。When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is less than the preset second threshold, level three is used as the event level.
具体地，在根据至少两种信息以及预设的等级划分表确定出事件等级的过程中，可先根据至少两种信息在预设的等级划分表中找出与所有信息对应的信息等级，进而可对所有信息等级与预设的第一等级、预设的第二等级进行比较，并根据比较结果确定出事件等级。Specifically, in the process of determining the event level based on at least two types of information and a preset level classification table, the information level corresponding to all information can be found in the preset level classification table based on at least two types of information, and then all information levels can be compared with the preset first level and the preset second level, and the event level can be determined based on the comparison results.
可以理解的是，比较方式可以但不局限于为：比较每个信息等级与预设的第一等级、预设的第二等级是否相同，并将与预设的第一等级以及预设的第二等级相同的信息的种类数量记作第一数量以及第二数量。在根据比较结果确定出事件等级的过程中，可根据第一数量与预设的第二阈值的大小关系以及第二数量与预设的第二阈值的大小关系确定出事件等级，其方式可以但不局限于为：当检测到第一数量大于或等于预设的第二阈值时，代表所有信息中与预设的第一等级相同的信息等级较多，此时可将一级作为事件等级；当检测到第一数量小于预设的第二阈值且第二数量大于或等于预设的第二阈值时，代表所有信息中与预设的第一等级相同的信息等级较少，与预设的第二等级相同的信息等级较多，此时可将二级作为事件等级；当检测到第二数量小于预设的第二阈值时，代表所有信息中与预设的第一等级相同的信息等级以及与预设的第二等级相同的信息等级均较少，此时可将三级作为事件等级。It is understandable that the comparison method may be, but is not limited to, comparing whether each information level is the same as the preset first level and the preset second level, and recording the number of types of information that are the same as the preset first level and the preset second level as the first number and the second number. In the process of determining the event level according to the comparison result, the event level may be determined according to the magnitude relationship between the first number and the preset second threshold value and the magnitude relationship between the second number and the preset second threshold value, and the method may be, but is not limited to, when it is detected that the first number is greater than or equal to the preset second threshold value, it means that there are more information levels that are the same as the preset first level in all information, and at this time, the first level may be used as the event level; when it is detected that the first number is less than the preset second threshold value and the second number is greater than or equal to the preset second threshold value, it means that there are fewer information levels that are the same as the preset first level in all information, and more information levels that are the same as the preset second level, and at this time, the second level may be used as the event level; when it is detected that the second number is less than the preset second threshold value, it means that there are fewer information levels that are the same as the preset first level and fewer information levels that are the same as the preset second level in all information, and at this time, the third level may be used as the event level.
此处，若预设的第二阈值为2，预设的第一等级为A，预设的第二等级为B，其他等级为C，则当所有的信息等级为A、A、B、A时，由于A的数量大于2，因此可将一级作为事件等级；当所有的信息等级为A、B、B、C时，由于A的数量小于2，B的数量等于2，因此可将二级作为事件等级；当所有的信息等级为A、B、C、C时，由于A与B的数量均小于2，因此可将三级作为事件等级。Here, if the preset second threshold is 2, the preset first level is A, the preset second level is B, and the other level is C, then when all information levels are A, A, B, A, since the number of A is greater than 2, the first level can be used as the event level; when all information levels are A, B, B, C, since the number of A is less than 2 and the number of B is equal to 2, the second level can be used as the event level; when all information levels are A, B, C, C, since the number of A and B are both less than 2, the third level can be used as the event level.
需要说明的是，预设的第二阈值可以但不局限于取信息等级的数量的一半。It should be noted that the preset second threshold may be, but is not limited to, half of the number of information levels.
作为本申请实施例的一种可选，按照与事件等级对应的发送方式，将事件描述发送至用户客户端，包括：As an option of the embodiment of the present application, sending the event description to the user client in a sending manner corresponding to the event level includes:
当识别出事件等级为一级时，按照预设的第一时间间隔以及电话呼叫方式，将事件描述发送至用户客户端；When the event level is identified as level one, the event description is sent to the user client according to a preset first time interval and telephone call method;
当识别出事件等级为二级时，按照预设的第二时间间隔以及语音助手播报方式，将事件描述发送至用户客户端；其中，第二时间间隔大于第一时间间隔；When the event level is identified as level 2, the event description is sent to the user client according to the preset second time interval and voice assistant broadcast mode; wherein the second time interval is greater than the first time interval;
当识别出事件等级为三级时，按照预设的第三时间间隔以及电子邮件通知方式，将事件描述发送至用户客户端；其中，第三时间间隔大于第二时间间隔。When the event level is identified as level three, the event description is sent to the user client according to a preset third time interval and email notification method; wherein the third time interval is greater than the second time interval.
具体地，在按照与事件等级对应的发送方式将事件描述发送至用户客户端的过程中，可先对事件等级进行识别，进而根据识别结果执行发送事件描述的操作，其过程可以但不局限于为：当识别出事件等级为一级时，表明事态紧急，需要尽快让用户知晓，因而可按照预设的第一时间间隔（如10秒）并通过电话呼叫的方式将事件描述发送给用户；当识别出事件等级为二级时，表明事态处于可控范围内，但急需采取应对措施，因而可按照预设的第二时间间隔（如3分钟）并通过语音助手播报的方式将事件描述发送给用户；当识别出事件等级为三级时，表明事态可控且无需立即处理，因而可按照预设的第三时间间隔（如20分钟）并通过电子邮件通知的方式将事件描述发送给用户。需要说明的是，预设的第三时间间隔大于预设的第二时间间隔，预设的第二时间间隔大于预设的第一时间间隔。Specifically, in the process of sending the event description to the user client in accordance with the sending method corresponding to the event level, the event level can be first identified, and then the operation of sending the event description is performed according to the identification result. The process can be but not limited to: when the event level is identified as level one, it indicates that the situation is urgent and needs to be made known to the user as soon as possible, so the event description can be sent to the user at a preset first time interval (such as 10 seconds) and by phone call; when the event level is identified as level two, it indicates that the situation is within the controllable range, but it is urgent to take countermeasures, so the event description can be sent to the user at a preset second time interval (such as 3 minutes) and by voice assistant broadcast; when the event level is identified as level three, it indicates that the situation is controllable and does not need to be handled immediately, so the event description can be sent to the user at a preset third time interval (such as 20 minutes) and by email notification. It should be noted that the preset third time interval is greater than the preset second time interval, and the preset second time interval is greater than the preset first time interval.
请参阅图3，图3示出了本申请实施例提供的又一种基于智能分析设备与业务模型的告警识别方法的整体流程图。Please refer to FIG. 3 , which shows an overall flow chart of another alarm identification method based on intelligent analysis equipment and business models provided in an embodiment of the present application.
如图3所示，该方法主要涉及到两种设备（感知探测设备与智能分析设备）以及两个终端（中心平台以及用户客户端）。As shown in FIG3 , the method mainly involves two devices (a perception detection device and an intelligent analysis device) and two terminals (a central platform and a user client).
可以理解的是，感知探测设备可以但不局限于为单个设备以及由多个设备协同组成，主要用于将采集到的可能存在入侵情况的初始图像或初始视频传输给中心平台；智能分析设备主要用于接收从中心平台发来的部分初始图像或部分初始视频，并对其进行复核分析处理，筛选出存在目标的图像或视频作为告警图像或告警视频传输给中心平台；中心平台主要用于在接收到由感知探测设备发送的初始图像或初始视频之后，根据业务需求或设备的精度限制等因素，筛选出需要进行复核分析的部分初始图像或部分初始视频，并将该部分初始图像或部分初始视频发送给智能分析设备，以确认这部分图像或视频中是否存在入侵目标，进而可接收经过智能分析设备确认的、存在目标的告警图像或告警视频并提取其中的信息，从而通过预设的业务模型将该信息转换为相应的事件描述发送至用户客户端。It can be understood that the perception detection device can be but not limited to a single device and can be composed of multiple devices working together, and is mainly used to transmit the collected initial images or initial videos that may contain intrusions to the central platform; the intelligent analysis device is mainly used to receive part of the initial images or part of the initial videos sent from the central platform, and review and analyze them, and screen out images or videos with targets as alarm images or alarm videos for transmission to the central platform; the central platform is mainly used to, after receiving the initial images or initial videos sent by the perception detection device, screen out part of the initial images or part of the initial videos that need to be reviewed and analyzed according to business requirements or factors such as equipment accuracy limitations, and send the part of the initial images or part of the initial videos to the intelligent analysis device to confirm whether there is an intrusion target in this part of the image or video, and then receive the alarm image or alarm video with the target confirmed by the intelligent analysis device and extract the information therein, so as to convert the information into a corresponding event description through a preset business model and send it to the user client.
请参阅图4，图4示出了本申请实施例提供的一种基于智能分析设备与业务模型的告警识别装置的结构示意图。Please refer to FIG. 4 , which shows a schematic diagram of the structure of an alarm identification device based on intelligent analysis equipment and a business model provided in an embodiment of the present application.
如图4所示，该基于智能分析设备与业务模型的告警识别装置至少可以包括第一处理模块401、第二处理模块402以及第三处理模块403，其中：As shown in FIG4 , the alarm identification device based on intelligent analysis equipment and business model may include at least a first processing module 401, a second processing module 402 and a third processing module 403, wherein:
第一处理模块401，用于当接收到由智能分析设备发送的至少一个告警图像时，对告警图像进行信息提取处理，得到至少两种信息；The first processing module 401 is used to extract information from the warning image when receiving at least one warning image sent by the intelligent analysis device to obtain at least two types of information;
第二处理模块402，用于基于至少两种信息以及预设的业务模型，得到事件描述；其中，预设的业务模型由至少两种样本信息以及与样本信息对应的事件描述训练得到；The second processing module 402 is used to obtain an event description based on at least two types of information and a preset business model; wherein the preset business model is trained by at least two types of sample information and event descriptions corresponding to the sample information;
第三处理模块403，用于基于至少两种信息以及预设的等级划分表，得到事件等级，并按照与事件等级对应的发送方式，将事件描述发送至用户客户端；其中，等级划分表至少包括两种信息以及与信息对应的等级。The third processing module 403 is used to obtain the event level based on at least two types of information and a preset level classification table, and send the event description to the user client in a sending method corresponding to the event level; wherein the level classification table includes at least two types of information and levels corresponding to the information.
在一些可能的实施例中，在接收到由智能分析设备发送的至少一个告警图像之前，还包括：In some possible embodiments, before receiving at least one warning image sent by the intelligent analysis device, the method further includes:
第一处理模块401具体用于：The first processing module 401 is specifically used for:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型属于预设的类型范围时，将初始图像发送给智能分析设备，以由智能分析设备对初始图像进行筛选处理并得到至少一个告警图像；其中，预设的类型范围至少包括两种感知探测设备的类型。When it is identified that the type of the perception detection device belongs to a preset type range, the initial image is sent to the intelligent analysis device, so that the intelligent analysis device screens and processes the initial image and obtains at least one alarm image; wherein the preset type range includes at least two types of perception detection devices.
在一些可能的实施例中，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：In some possible embodiments, before obtaining the event description based on the at least two types of information and the preset business model, the method further includes:
第一处理模块401具体用于：The first processing module 401 is specifically used for:
当接收到由感知探测设备发送的初始图像时，获取感知探测设备的类型；When receiving an initial image sent by a perception detection device, obtaining a type of the perception detection device;
当识别出感知探测设备的类型不属于预设的类型范围时，对初始图像进行信息提取处理，得到至少两种信息。When it is identified that the type of the perception detection device does not belong to the preset type range, information extraction processing is performed on the initial image to obtain at least two types of information.
在一些可能的实施例中，在基于至少两种信息以及预设的业务模型，得到事件描述之前，还包括：In some possible embodiments, before obtaining the event description based on the at least two types of information and the preset business model, the method further includes:
第一处理模块401具体用于：The first processing module 401 is specifically used for:
当接收到由智能分析设备发送的至少一个告警视频时，获取告警视频中任意两个相邻帧图像；When at least one alarm video sent by the intelligent analysis device is received, any two adjacent frame images in the alarm video are obtained;
计算任意两个相邻帧图像之间的像素差值；Calculate the pixel difference between any two adjacent frame images;
当检测到任意一个像素差值大于预设的第一阈值时，对像素差值所对应的后帧图像进行信息提取处理，得到至少两种信息；其中，后帧图像为任意两个相邻帧图像中在时间上后出现的一帧图像。When it is detected that any pixel difference is greater than a preset first threshold, information extraction processing is performed on the subsequent frame image corresponding to the pixel difference to obtain at least two types of information; wherein the subsequent frame image is a frame image that appears later in time between any two adjacent frame images.
在一些可能的实施例中，至少两种信息包括行为信息以及区域信息；In some possible embodiments, the at least two types of information include behavior information and area information;
基于至少两种信息以及预设的业务模型，得到事件描述，包括：Based on at least two types of information and a preset business model, an event description is obtained, including:
第二处理模块402具体用于：The second processing module 402 is specifically used for:
当检测到至少两种信息中存在行为信息时，获取区域信息的重要等级，并根据重要等级确定出描述顺序；When it is detected that there is behavior information in at least two types of information, obtaining the importance level of the area information, and determining the description order according to the importance level;
将描述顺序、行为信息以及区域信息输入至预设的业务模型，得到事件描述。The description order, behavior information, and region information are input into a preset business model to obtain an event description.
在一些可能的实施例中，基于至少两种信息以及预设的等级划分表，得到事件等级，包括：In some possible embodiments, the event level is obtained based on at least two types of information and a preset level classification table, including:
第三处理模块403具体用于：The third processing module 403 is specifically used for:
基于预设的等级划分表，得到与每种信息对应的信息等级；Based on a preset level classification table, obtain the information level corresponding to each type of information;
当检测到信息等级与预设的第一等级相同的信息的种类数量大于或等于预设的第二阈值时，将一级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is greater than or equal to a preset second threshold, the first level is taken as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量大于或等于预设的第二阈值时，将二级作为事件等级；When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is greater than or equal to the preset second threshold, the second level is used as the event level;
当检测到信息等级与预设的第一等级相同的信息的种类数量小于预设的第二阈值，且信息等级与预设的第二等级相同的信息的种类数量小于预设的第二阈值时，将三级作为事件等级。When it is detected that the number of types of information with the same information level as the preset first level is less than the preset second threshold, and the number of types of information with the same information level as the preset second level is less than the preset second threshold, level three is used as the event level.
在一些可能的实施例中，按照与事件等级对应的发送方式，将事件描述发送至用户客户端，包括：In some possible embodiments, the event description is sent to the user client in a sending manner corresponding to the event level, including:
第三处理模块403具体用于：The third processing module 403 is specifically used for:
当识别出事件等级为一级时，按照预设的第一时间间隔以及电话呼叫方式，将事件描述发送至用户客户端；When the event level is identified as level one, the event description is sent to the user client according to a preset first time interval and telephone call method;
当识别出事件等级为二级时，按照预设的第二时间间隔以及语音助手播报方式，将事件描述发送至用户客户端；其中，第二时间间隔大于第一时间间隔；When the event level is identified as level 2, the event description is sent to the user client according to a preset second time interval and voice assistant broadcast mode; wherein the second time interval is greater than the first time interval;
当识别出事件等级为三级时，按照预设的第三时间间隔以及电子邮件通知方式，将事件描述发送至用户客户端；其中，第三时间间隔大于第二时间间隔。When the event level is identified as level three, the event description is sent to the user client according to a preset third time interval and email notification method; wherein the third time interval is greater than the second time interval.
请参阅图5，图5示出了本申请实施例提供的又一种基于智能分析设备与业务模型的告警识别装置的结构示意图。Please refer to FIG. 5 , which shows a schematic diagram of the structure of another alarm identification device based on intelligent analysis equipment and business models provided in an embodiment of the present application.
如图5所示，该基于智能分析设备与业务模型的告警识别装置500可以包括至少一个处理器501、至少一个网络接口504、用户接口503、存储器505以及至少一个通信总线502。As shown in FIG. 5 , the alarm identification device 500 based on intelligent analysis equipment and business models may include at least one processor 501 , at least one network interface 504 , a user interface 503 , a memory 505 and at least one communication bus 502 .
其中，通信总线502可用于实现上述各个组件的连接通信。The communication bus 502 may be used to realize the connection and communication among the above-mentioned components.
其中，用户接口503可以包括按键，可选用户接口还可以包括标准的有线接口、无线接口。The user interface 503 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.
其中，网络接口504可以但不局限于包括蓝牙模块、NFC模块、Wi-Fi模块等。The network interface 504 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.
其中，处理器501可以包括一个或者多个处理核心。处理器501利用各种接口和线路连接整个基于智能分析设备与业务模型的告警识别装置500内的各个部分，通过运行或执行存储在存储器505内的指令、程序、代码集或指令集，以及调用存储在存储器505内的数据，执行基于智能分析设备与业务模型的告警识别装置500的各种功能和处理数据。可选的，处理器501可以采用DSP、FPGA、PLA中的至少一种硬件形式来实现。处理器501可集成CPU、GPU和调制解调器等中的一种或几种的组合。其中，CPU主要处理操作系统、用户界面和应用程序等；GPU用于负责显示屏所需要显示的内容的渲染和绘制；调制解调器用于处理无线通信。可以理解的是，上述调制解调器也可以不集成到处理器501中，单独通过一块芯片进行实现。Among them, the processor 501 may include one or more processing cores. The processor 501 uses various interfaces and lines to connect the various parts of the entire alarm identification device 500 based on intelligent analysis equipment and business models, and executes various functions and processes data of the alarm identification device 500 based on intelligent analysis equipment and business models by running or executing instructions, programs, code sets or instruction sets stored in the memory 505, and calling data stored in the memory 505. Optionally, the processor 501 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 501 can integrate one or a combination of CPU, GPU, modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 501, and it can be implemented by a single chip.
其中，存储器505可以包括RAM，也可以包括ROM。可选的，该存储器505包括非瞬时性计算机可读介质。存储器505可用于存储指令、程序、代码、代码集或指令集。存储器505可包括存储程序区和存储数据区，其中，存储程序区可存储用于实现操作系统的指令、用于至少一个功能的指令（比如触控功能、声音播放功能、图像播放功能等）、用于实现上述各个方法实施例的指令等；存储数据区可存储上面各个方法实施例中涉及到的数据等。存储器505可选的还可以是至少一个位于远离前述处理器501的存储装置。如图5所示，作为一种计算机存储介质的存储器505中可以包括操作系统、网络通信模块、用户接口模块以及基于智能分析设备与业务模型的告警识别应用程序。Among them, the memory 505 may include RAM or ROM. Optionally, the memory 505 includes a non-transitory computer-readable medium. The memory 505 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 505 may also be at least one storage device located away from the aforementioned processor 501. As shown in Figure 5, the memory 505 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an alarm identification application based on an intelligent analysis device and a business model.
具体地，处理器501可以用于调用存储器505中存储的基于智能分析设备与业务模型的告警识别应用程序，并具体执行以下操作：Specifically, the processor 501 may be used to call the alarm identification application based on the intelligent analysis device and the business model stored in the memory 505, and specifically perform the following operations:
当接收到由智能分析设备发送的至少一个告警图像时，对告警图像进行信息提取处理，得到至少两种信息；When receiving at least one warning image sent by the intelligent analysis device, performing information extraction processing on the warning image to obtain at least two types of information;
基于至少两种信息以及预设的业务模型，得到事件描述；其中，预设的业务模型由至少两种样本信息以及与样本信息对应的事件描述训练得到；Based on at least two kinds of information and a preset business model, an event description is obtained; wherein the preset business model is trained by at least two kinds of sample information and event descriptions corresponding to the sample information;
基于至少两种信息以及预设的等级划分表，得到事件等级，并按照与事件等级对应的发送方式，将事件描述发送至用户客户端；其中，等级划分表至少包括两种信息以及与信息对应的等级。Based on at least two types of information and a preset level classification table, an event level is obtained, and the event description is sent to the user client in a sending mode corresponding to the event level; wherein the level classification table includes at least two types of information and levels corresponding to the information.
本申请还提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现上述方法的步骤。其中，计算机可读存储介质可以包括但不限于任何类型的盘，包括软盘、光盘、DVD、CD-ROM、微型驱动器以及磁光盘、ROM、RAM、EPROM、EEPROM、DRAM、VRAM、闪速存储器设备、磁卡或光卡、纳米系统（包括分子存储器IC），或适合于存储指令和/或数据的任何类型的媒介或设备。The present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and/or data.
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
在本申请所提供的几个实施例中，应该理解到，所揭露的装置，可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，例如单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，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。另一点，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服务接口，装置或单元的间接耦合或通信连接，可以是电性或其它的形式。In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.
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 distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
另外，在本申请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 application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
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，可以存储在一个计算机可读取存储器中。基于这样的理解，本申请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器中，包括若干指令用以使得一台计算机设备（可为个人计算机、服务器或者网络设备等）执行本申请各个实施例方法的全部或部分步骤。而前述的存储器包括：U盘、只读存储器（Read-Only Memory， ROM）、随机存取存储器（RandomAccess Memory，RAM）、移动硬盘、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 implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM), random access memory (RAM), mobile hard disk, disk or optical disk and other media that can store program codes.
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通进程序来指令相关的硬件来完成，该程序可以存储于一计算机可读存储器中，存储器可以包括：闪存盘、只读存储器（Read-Only Memory， ROM）、随机存取器（Random AccessMemory，RAM）、磁盘或光盘等。A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by entering a program to instruct related hardware. The program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.
以上者，仅为本公开的示例性实施例，不能以此限定本公开的范围。即但凡依本公开教导所作的等效变化与修饰，皆仍属本公开涵盖的范围内。本领域技术人员在考虑说明书及实践这里的公开后，将容易想到本公开的其实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未记载的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本公开的范围和精神由权利要求限定。The above are only exemplary embodiments of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410560564.4ACN118570521A (en) | 2024-05-08 | 2024-05-08 | A method and device for identifying alarms based on intelligent analysis equipment and business models |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410560564.4ACN118570521A (en) | 2024-05-08 | 2024-05-08 | A method and device for identifying alarms based on intelligent analysis equipment and business models |
| Publication Number | Publication Date |
|---|---|
| CN118570521Atrue CN118570521A (en) | 2024-08-30 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410560564.4APendingCN118570521A (en) | 2024-05-08 | 2024-05-08 | A method and device for identifying alarms based on intelligent analysis equipment and business models |
| Country | Link |
|---|---|
| CN (1) | CN118570521A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119625961A (en)* | 2025-02-11 | 2025-03-14 | 宁波智蜂物联有限公司 | Multimodal intelligent security method and system driven by brain-like edge computing |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119625961A (en)* | 2025-02-11 | 2025-03-14 | 宁波智蜂物联有限公司 | Multimodal intelligent security method and system driven by brain-like edge computing |
| Publication | Publication Date | Title |
|---|---|---|
| CN106203458B (en) | Crowd video analysis method and system | |
| CN112364696B (en) | Method and system for improving family safety by utilizing family monitoring video | |
| CN110852147B (en) | Security alarm method, security alarm device, server and computer readable storage medium | |
| WO2020073505A1 (en) | Image processing method, apparatus and device based on image recognition, and storage medium | |
| US20130216107A1 (en) | Method of surveillance by face recognition | |
| CN109993044B (en) | Telecommunications fraud identification system, method, apparatus, electronic device, and storage medium | |
| WO2017117879A1 (en) | Personal identification processing method, apparatus and system | |
| WO2021223665A1 (en) | Method and device for reducing invalid warnings | |
| CN111814725A (en) | An early warning method for judging surveillance video fire based on CNN+LSTM+MLP combined neural network | |
| JP6621092B1 (en) | Risk determination program and system | |
| CN112419639A (en) | Video information acquisition method and device | |
| CN118570521A (en) | A method and device for identifying alarms based on intelligent analysis equipment and business models | |
| CN117876956A (en) | Abnormal behavior detection method and device, electronic equipment and storage medium | |
| CN118351555A (en) | Campus anti-slush edge computing system and method based on semantic recognition and image recognition technology | |
| CN104050785A (en) | Safety alert method based on virtualized boundary and face recognition technology | |
| CN113342978A (en) | City event processing method and device | |
| CN115861929A (en) | Monitoring system, method, equipment and storage medium for service specification of electric power business hall | |
| CN114973135A (en) | A head-and-shoulders-based time-series video sleeping post identification method, system and electronic device | |
| US20220414377A1 (en) | System and method for presenting statements captured at an incident scene | |
| CN112347810A (en) | Method and device for detecting moving target object and storage medium | |
| CN112419638B (en) | Method and device for acquiring alarm video | |
| CN110782622A (en) | Safety monitoring system, safety detection method, safety detection device and electronic equipment | |
| CN118506808A (en) | Baling identification and processing method and device, electronic equipment and storage medium | |
| CN111563174A (en) | Image processing method, image processing apparatus, electronic device, and storage medium | |
| CN114241401B (en) | Abnormality determination method, device, equipment, medium and product |
| 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 |