A method and system for monitoring video data

Abstract

The application discloses a kind of processing method and system of monitoring video data, main purpose is to solve the problem of low efficiency of existing monitoring video data processing. Including: obtaining the job task information of each monitoring object and the monitoring video data matched with the monitoring object;Determine the job object and the job flow node based on the job task information, and at least two groups of image data to be processed are intercepted from the monitoring video data according to the job object, the job flow node;According to the monitoring target in the preset monitoring sequence, the image data is extracted to obtain the target feature, the pointer in the preset monitoring sequence is sequentially polled to different monitoring targets and is directed;The monitoring state of the monitoring video data is obtained based on the target feature and the monitoring whitelist of the job task information, to determine the monitoring result based on the monitoring state.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a processing method and system for monitoring video data. BACKGROUND

[0002] With the advent of the era of big data and artificial intelligence, more and more enterprises choose unmanned and fully automatic production methods, and replace manual work with fully automatic machine devices. In order to ensure the normal operation of the machine devices, a video monitoring method is usually used to monitor and warn a specific area to ensure the safety of production.

[0003] At present, the processing of monitoring video data is usually based on the analysis of image data and the determination of monitoring results based on image features. However, the number of images in the video data is too large, and if all the video data is analyzed and processed, the processing capacity of the system is greatly consumed, and a system with strong computing capacity needs to be configured online with the operating system, or a separate monitoring system needs to be configured, which seriously affects the processing efficiency of the operating data and increases the operating processing cost. Therefore, a processing method for monitoring video data is needed to solve the above problems. SUMMARY

[0004] Therefore, the present application provides a processing method and system for monitoring video data, which mainly aims to solve the problem of low processing efficiency of the existing monitoring video data.

[0005] According to one aspect of the present application, a processing method for monitoring video data is provided, comprising:

[0006] obtaining the job task information of each monitoring object and the monitoring video data matched with the monitoring object;

[0007] determining the job object and the job flow node based on the job task information, and intercepting at least two groups of image data to be processed from the monitoring video data according to the job object and the job flow node;

[0008] extracting the image data according to the monitoring target in the preset monitoring sequence to obtain the target feature, wherein the pointer in the preset monitoring sequence is used to poll and point to different monitoring targets in time sequence;

[0009] matching the target feature with the monitoring whitelist of the job task information to obtain the monitoring state of the monitoring video data, and determining the monitoring result based on the monitoring state.

[0010] According to another aspect of the present application, a processing system for monitoring video data is provided, comprising:

[0011] An acquisition module is configured to acquire job task information of each monitoring object and monitoring video data matched with the monitoring object;

[0012] An intercepting module is configured to determine a job object and a job flow node based on the job task information, and intercept at least two groups of image data to be processed from the monitoring video data according to the job object and the job flow node;

[0013] An extracting module is configured to extract the image data according to a monitoring target in a preset monitoring sequence to obtain a target feature, and a pointer in the preset monitoring sequence is used to poll different monitoring targets in a time sequence;

[0014] A matching module is configured to match the target feature with a monitoring whitelist of the job task information to obtain a monitoring state of the monitoring video data, and determine a monitoring result based on the monitoring state.

[0015] According to another aspect of the present application, a storage medium is provided, and the storage medium stores at least one executable instruction, and the executable instruction causes a processor to perform operations corresponding to the above-mentioned processing method of monitoring video data.

[0016] According to still another aspect of the present application, a terminal is provided, and the terminal comprises a processor, a memory, a communication interface and a communication bus, and the processor, the memory and the communication interface complete communication with each other through the communication bus.

[0017] The memory is configured to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the above-mentioned processing method of monitoring video data.

[0018] Through the above technical solutions, the technical solutions provided by the embodiments of the present application have at least the following advantages:

[0019] This invention provides a method and system for processing surveillance video data. Compared with existing technologies, this invention obtains the task information of each monitored object and the surveillance video data matching the monitored object; determines the task object and task flow node based on the task information, and extracts at least two sets of image data to be processed from the surveillance video data according to the task object and the task flow node; extracts the image data according to the monitoring target in the preset monitoring sequence to obtain target features, wherein the pointer in the preset monitoring sequence is a polling pointer to different monitoring targets in time sequence; matches the target features with the monitoring whitelist of the task information to obtain the monitoring status of the surveillance video data, and determines the monitoring result based on the monitoring status. This greatly reduces the processing pressure on video data, flexibly meets the needs of using video data as a monitoring basis in different scenarios, effectively improves the monitoring and processing capabilities of task data, and thus improves the processing efficiency of surveillance video data.

[0020] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0022] Figure 1 A flowchart of a method for processing surveillance video data provided by an embodiment of the present invention is shown;

[0023] Figure 2 This diagram illustrates a block diagram of a video surveillance data processing system provided by an embodiment of the present invention.

[0024] Figure 3 A schematic diagram of the structure of a terminal provided in an embodiment of the present invention is shown. Detailed Implementation

[0025] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0026] The embodiment of the present application provides a processing method of monitoring video data, as shown in the figure, the method comprises: Figure 1

[0027] 101、obtain the job task information of each monitoring object and the monitoring video data matched with the monitoring object.

[0028] In the embodiment of the present application, the current execution end is a service end for processing monitoring data, which can be a cloud server or a local server, so as to call the video data of the video shooting device installed at each monitoring node for processing. Wherein, the monitoring object is the content to be monitored, including but not limited to the product on the industrial automation assembly line, the equipment group in the communication equipment room, etc. Correspondingly, the job task information is the time, frequency, job equipment and other process content of the job of the monitoring object, so as to obtain the job object, job process node and other content based on the job task information analysis. In addition, the current execution end interacts with the monitoring device installed at each monitoring object, so as to obtain the monitoring video data matched with the monitoring object. At this time, the monitoring video data can be the video content collected according to the preset time interval, or the real-time collected video content, which is not limited in the embodiment of the present application.

[0029] 102、determine the job object and the job process node based on the job task information, and intercept at least two groups of image data to be processed from the monitoring video data according to the job object and the job process node.

[0030] In the embodiment of the present application, the current execution end analyzes the task information to determine the job object and the job process node. At this time, the job object is the specific object of the industrial job of the monitoring object, for example, the monitoring object is the product on the assembly line, and the job object is the installation operation of installing accessories on the product. For example, the monitoring object is the mechanical and electrical communication equipment, and the job object is the communication signal transmitted by the mechanical and electrical communication equipment, which is not limited in the embodiment of the present application. At the same time, the job process node is used to represent the process required to be executed by the monitoring object in the job process, for example, the monitoring object is the product on the assembly line, and the job process node can include two process nodes, which are placing accessories and welding connection points, which is not limited in the embodiment of the present application. In addition, in order to reduce the processing pressure of the video data, the current execution end intercepts at least two groups of image data to be processed from all the monitoring video according to the job object and the job process node. Wherein, each group of image data to be processed contains a plurality of frame video data within 1 second, so as to reduce the number of monitoring video data to be processed.

[0031] 103、extract the image data according to the monitoring target in the preset monitoring sequence to obtain the target feature.

[0032] ​In the embodiment of the present application, when processing image data, the current execution end extracts features of the image data according to a monitoring target in a preset monitoring sequence, and obtains target features. At this time, a pointer in the preset monitoring sequence is used to poll different monitoring targets in a time sequence, and the monitoring target is a monitoring requirement for a monitoring object or a job object, such as a job time of the monitoring object, a job position of the job object, and the like, which can be set based on different monitoring requirements, and the embodiment of the present application is not limited in this regard. Meanwhile, in the embodiment of the present application, in order to flexibly monitor the monitoring object, a plurality of monitoring targets are stored in the preset monitoring sequence, and are polled and pointed to by the pointer at a specific time interval, thereby improving the flexibility of monitoring.

[0033] 104. Matching the target features with a monitoring whitelist based on the job task information to obtain a monitoring state of the monitoring video data, so as to determine a monitoring result based on the monitoring state.

[0034] In the embodiment of the present application, after the target features are extracted, the current execution end compares the target features with white features in a monitoring whitelist corresponding to the job task information to determine a monitoring state of the monitoring video data, so as to determine a monitoring result of the final monitoring object based on the monitoring state. At this time, the monitoring whitelist stores normal image features for different job task information, which are matched with the target features as white features. At this time, the white features can be configured based on monitoring requirements, image recognition accuracy, and the like, and the embodiment of the present application is not limited in this regard.

[0035] In another embodiment of the present application, in order to further limit and illustrate, the step of determining a job object and a job flow node based on the job task information and intercepting at least two groups of to-be-processed image data from the monitoring video data according to the job object and the job flow node includes:

[0036] Extracting a job number and job equipment information in the job task information, and determining a job object and a job flow node corresponding to the job number and the job equipment information based on a preset job mapping relationship;

[0037] Determining a monitoring operation node, a monitoring time node, and a monitoring position node of the job object in the job flow node;

[0038] Filtering a plurality of video nodes from the monitoring video data based on the monitoring operation node, the monitoring time node, and the monitoring position node, and intercepting at least two groups of to-be-processed image data according to the video nodes.

[0039] In order to accurately steal the image data to be processed, the current execution end first extracts the job number in the job task information and the job equipment information. In the embodiment of the present application, the user will mark different job contents based on the pre-set job number when creating the job task information, and correspondingly, the job equipment. Therefore, the generated job task information contains the job number and the job equipment information. For example, the generated job task information contains the number 001 of product a and the number 002 of product b. The job equipment information corresponding to the number 001 is automation equipment 1 and automation equipment 2. The job equipment information corresponding to the number is automation equipment 2. The embodiment of the present application is not limited. Further, the current execution end determines the job object and the job process node corresponding to the job number and the job equipment information based on the pre-set job mapping relationship. At this time, the pre-set job mapping relationship includes the mapping relationship between different job numbers, different job equipment information and different job objects, different job process nodes, so as to determine the job object and the job process node.

[0040] It should be noted that in the embodiment of the present application, after the job object is determined, in order to accurately obtain the data that needs to be monitored and processed, the current execution end determines the monitoring operation node, the monitoring time node and the monitoring position node of the job object in the job process node. That is, it can be determined by manual selection, or it can be determined based on the pre-configured monitoring requirements of the job object. The monitoring operation node is used to represent the node that needs to be monitored when the job object is in the job process node. The monitoring time node is used to represent the specified time when the job object is in the job process node. The monitoring position node is used to represent the fixed position that needs to be monitored when the job object is on the operation equipment. The embodiment of the present application is not limited. Further, according to the determined monitoring operation node, monitoring time node and monitoring position node, the video node is screened from the monitoring video data. At this time, the video content of the target monitoring device or part of the video captured can be determined based on the time, position and the like captured by each monitoring device to belong to the aforementioned monitoring operation node, monitoring time node and monitoring position node, so as to intercept at least two groups of image data according to the video node. The embodiment of the present application is not limited.

[0041] In another embodiment of the present application, in order to further limit and illustrate, the step of screening a plurality of video nodes from the monitoring video data based on the monitoring operation node, the monitoring time node and the monitoring position node includes:

[0042] searching all first reference video nodes based on the monitoring time node, and calling the second reference video node corresponding to the target monitoring device in the first reference video node according to the monitoring position node;

[0043] determine an operation feature corresponding to the monitoring operation node, and screen a plurality of video nodes from the second reference video node according to the operation feature.

[0044] In order to accurately screen a plurality of video nodes, reduce the processing pressure of monitoring image data, and improve the processing efficiency of monitoring data, when screening the video nodes, the current execution end first searches all first reference video nodes based on a monitoring time node, for example, finds the video nodes according to a monitoring time node of every 5 minutes, and calls the second reference video node matched with the target monitoring device according to a monitoring position node. At this time, different monitoring position nodes are each installed with a device for shooting different monitoring devices, and the embodiment of the application does not make specific limitation. At this time, the obtained reference video node generally includes all frame video data of a time unit, for example, a reference video node is a shooting node of one second, and each reference video node includes all frame image data within 1 second, and the embodiment of the application does not make specific limitation.

[0045] When the second reference video node is determined, in order to extract the corresponding image according to the feature, therefore, the current execution end first determines the operation feature of the monitoring operation node, for example, a welding operation, and the corresponding operation feature is a welding gun, which can be determined based on a preconfigured feature corresponding relationship, and the embodiment of the application does not make specific limitation. When the operation feature is determined, a plurality of video nodes are screened from the second reference video node according to the operation feature, that is, the image data in the second reference video node is pre-screened according to the operation feature. The screening can be performed based on the similarity of the image features, and the video node greater than a similarity threshold value is regarded as the final screened video node, and the embodiment of the application does not make specific limitation.

[0046] In another embodiment of the application, in order to further limit and illustrate, the step of intercepting at least two groups of image data to be processed according to the video node includes:

[0047] Compare the image similarity between a plurality of image data corresponding to the video node and the pixel range of the operation feature.

[0048] Intercept at least two groups of image data to be processed from the image data based on the comparison result of the image similarity and the pixel range.

[0049] In order to reduce the processing pressure of image data and achieve efficient and accurate monitoring, in the embodiment of the present application, when at least two groups of image data to be processed are intercepted, specifically, the image similarity between all image data in the video node and the pixel range of the operation feature are compared. Since the video node contains multiple image data, the similarity between each image data can be calculated in turn, and the pixel range of each operation feature is compared at the same time to find the most accurate image data. In addition, when the current execution end intercepts two groups of image data to be processed, image similarity sorting and pixel range sorting can be performed, and the image similarity and pixels of the two groups with the highest similarity are used as the image data to be processed. At this time, each group of image data includes all frame images in a node, that is, one group of image data to be processed is used to represent all frame image data in a video node, which is not limited in the embodiment of the present application.

[0050] In another embodiment of the present application, in order to further limit and illustrate, the step of extracting the image data according to the monitoring target in the preset monitoring sequence to obtain the target feature includes:

[0051] The preset monitoring sequence is called, and the pointer pointing in the preset monitoring sequence is queried to determine the monitoring target corresponding to the pointer pointing;

[0052] The image data is identified based on the specially trained identification model to obtain the target feature.

[0053] In order to extract the features in the image data and then identify and monitor based on the extracted target features, the current execution end first calls the preset monitoring sequence. At this time, the pointer in the preset monitoring sequence points to a monitoring target at a time interval, for example, there are 10 monitoring targets in the monitoring sequence, and the pointer starts from the first monitoring target at a time interval of 0 seconds. When the pointer is queried to point to the third monitoring target at 30 seconds, the embodiment of the present application is not limited. Further, when the monitoring target is determined, the image data is identified based on the feature identification module. At this time, the feature identification module is obtained by training the feature training sample set corresponding to the queried monitoring target, for example, when the monitoring target is a welding operation position, the feature extraction model is trained based on the feature training sample set of the welding operation position. The feature extraction model can be a machine learning model such as neural network, convolutional neural network, etc., which is not limited in the embodiment of the present application.

[0054] In another embodiment of the present application, in order to further limit and illustrate, the step of matching the target feature with the monitoring white list of the operation task information to obtain the monitoring state of the monitoring video data includes:

[0055] acquire a monitoring whitelist of the job task information;

[0056] when the feature similarity between the target feature and the security feature is greater than a preset threshold, determining that the monitoring state of the monitoring video data is a normal monitoring state;

[0057] when the feature similarity between the target feature and the security feature is less than or equal to a preset threshold, determining that the monitoring state of the monitoring video data is an abnormal monitoring state.

[0058] In order to accurately obtain the monitoring state based on the target feature, determine the monitoring result, and improve the monitoring accuracy, the current execution end first acquires the monitoring whitelist of the job task information, and the monitoring whitelist includes the pre-recorded security feature, so as to compare with the target feature. Among them, different job task information can match different monitoring whitelists. When the feature similarity between the target feature and the security feature is greater than a preset threshold, it means that the job situation corresponding to the target feature is normal at this time, therefore, the monitoring state of the monitoring video data is determined as a normal monitoring state. Correspondingly, when the feature similarity between the target feature and the security feature is less than or equal to a preset threshold, it means that the job situation corresponding to the target feature is abnormal at this time, therefore, the monitoring state of the monitoring video data is determined as an abnormal monitoring state. The preset threshold is configured based on the monitoring demand, and the embodiment of the application does not make specific limitation.

[0059] In another embodiment of the application, in order to further limit and illustrate, the step further comprises:

[0060] counting the number of states of the normal monitoring state or the abnormal monitoring state of the monitoring video data according to the monitoring task information corresponding to the job task information;

[0061] if the number of states exceeds a preset number threshold, determining that the monitoring result is a monitoring warning, and sending a warning information;

[0062] if the number of states does not exceed a preset number threshold, determining that the monitoring result is a monitoring continuation, and updating the monitoring video data.

[0063] In order to meet the accuracy requirement of monitoring, the current execution end counts the normal monitoring state and the abnormal monitoring state according to the monitoring task information corresponding to the job task information, and respectively obtains the corresponding state number. At this time, the preset number threshold is configured in the monitoring task information based on the monitoring requirement (including the requirement of counting the number of normal monitoring state or abnormal monitoring state). In another implementation scenario, when the monitoring is strict, the monitoring task information records the counting of the abnormal monitoring state, and the corresponding preset number threshold is configured for the number of abnormal monitoring state. For example, when the number of abnormal monitoring state exceeds 10, and no matter how many the number of normal monitoring state is, the monitoring result is determined as monitoring warning. In another implementation scenario, when the monitoring is loose, the monitoring task information records the counting of the normal monitoring state, and the corresponding preset number threshold is configured for the number of normal monitoring state. For example, when the number of normal monitoring state is less than 100, and the number of abnormal monitoring state exceeds 50, the monitoring result is determined as monitoring warning. Correspondingly, if the state number exceeds the preset number threshold, the monitoring result is determined as monitoring persistence, and the monitoring video data captured by each monitoring device is updated to re-execute the steps of determining the job object and the job process node based on the job task information, and intercepting at least two groups of image data to be processed from the monitoring video data according to the job object and the job process node.

[0064] In the embodiment of the application, a monitoring video data processing method is provided. Compared with the prior art, the embodiment of the application acquires job task information of each monitoring object and monitoring video data matched with the monitoring object; determines a job object and a job process node based on the job task information, and intercepts at least two groups of image data to be processed from the monitoring video data according to the job object and the job process node; extracts the image data according to a monitoring target in a preset monitoring sequence to obtain a target feature, wherein the pointer in the preset monitoring sequence is a polling pointer for different monitoring targets in time sequence; matches the target feature with a monitoring white list in the job task information to obtain a monitoring state of the monitoring video data, and determines a monitoring result based on the monitoring state, thereby greatly reducing the processing pressure of the video data, flexibly meeting the requirement of using video data as monitoring basis in different scenes, effectively improving the monitoring processing capability of the job data, and thereby improving the processing efficiency of the monitoring video data.

[0065] Further, as an implementation of the method shown in the above Figure 1 , the embodiment of the application provides a monitoring video data processing system, as shown in the Figure 2 , which comprises:

[0066] The acquisition module 21 is used to acquire the job task information of each monitored object and the monitoring video data matched with the monitored object;

[0067] The interception module 22 is used to determine the task object and task process node based on the task information, and to intercept at least two sets of image data to be processed from the monitoring video data according to the task object and the task process node.

[0068] Extraction module 23 is used to extract the image data according to the monitoring targets in the preset monitoring sequence to obtain target features, wherein the pointer in the preset monitoring sequence is to poll different monitoring targets in time sequence;

[0069] The matching module is used to match the target features with the monitoring whitelist of the job task information to obtain the monitoring status of the monitoring video data, and to determine the monitoring result based on the monitoring status.

[0070] This invention provides a system for processing surveillance video data. Compared with existing technologies, this invention acquires task information for each monitored object and surveillance video data matching the monitored object; determines the task object and task flow nodes based on the task information, and extracts at least two sets of image data to be processed from the surveillance video data according to the task object and the task flow nodes; extracts target features from the image data according to the monitoring targets in a preset monitoring sequence, where the pointers in the preset monitoring sequence poll different monitoring targets in a time sequence; matches the target features with the monitoring whitelist of the task information to obtain the monitoring status of the surveillance video data, and determines the monitoring result based on the monitoring status. This greatly reduces the processing pressure on video data, flexibly meets the needs of using video data as a monitoring basis in different scenarios, effectively improves the monitoring and processing capabilities of task data, and thus improves the processing efficiency of surveillance video data.

[0071] According to one embodiment of the present invention, a storage medium is provided, the storage medium storing at least one executable instruction, which can perform the method for processing surveillance video data in any of the above method embodiments.

[0072] Figure 3 The diagram shows a structural schematic of a terminal according to an embodiment of the present invention. The specific implementation of the terminal is not limited by the specific embodiments of the present invention.

[0073] like Figure 3As shown, the terminal can include a processor 302, a communications interface 304, a memory 306, and a communications bus 308.

[0074] The processor 302, the communications interface 304, and the memory 306 can communicate with each other through the communications bus 308.

[0075] The communications interface 304 is configured to communicate with network elements such as clients or other servers.

[0076] The processor 302 is configured to execute the program 310, and in particular, can execute the related steps in the processing method embodiments of the monitored video data.

[0077] In particular, the program 310 can include program codes including computer operation instructions.

[0078] The processor 302 can be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present application. The one or more processors included in the terminal can be the same type of processors, such as one or more CPUs; or can be different types of processors, such as one or more CPUs and one or more ASICs.

[0079] The memory 306 is configured to store the program 310. The memory 306 can include a high-speed RAM memory, and can also include a non-volatile memory such as at least one disk memory.

[0080] The program 310 can be specifically configured to cause the processor 302 to perform the following operations:

[0081] Obtain job task information of each monitoring object and monitoring video data matched with the monitoring object;

[0082] Determine a job object and a job flow node based on the job task information, and intercept at least two groups of image data to be processed from the monitoring video data according to the job object and the job flow node;

[0083] Extract the image data according to a monitoring target in a preset monitoring sequence to obtain a target feature, wherein a pointer in the preset monitoring sequence is a polling pointer to different monitoring targets in a time sequence;

[0084] The target feature is matched with the monitoring white list of the job task information to obtain a monitoring state of the monitoring video data, and a monitoring result is determined based on the monitoring state.

[0085] It is apparent that those skilled in the art should understand that the modules or steps of the present application described above can be realized by a general computing system, which can be centralized on a single computing system or distributed on a network composed of multiple computing systems, and optionally, they can be realized by program codes executable by a computing system, so that they can be stored in a storage system and executed by a computing system, and in some cases, the steps shown or described can be executed in different order, or they can be respectively manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module to realize. Thus, the present application is not limited to any specific combination of hardware and software.

[0086] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for processing surveillance video data, characterized in that, include: Obtain the task information of each monitored object and the monitoring video data that matches the monitored object; Based on the task information, the task object and task process node are determined, and at least two sets of image data to be processed are extracted from the monitoring video data according to the task object and the task process node. The image data is extracted according to the monitoring targets in the preset monitoring sequence to obtain target features. The pointer in the preset monitoring sequence is to poll different monitoring targets in time sequence. The monitoring status of the monitoring video data is obtained by matching the target features with the monitoring whitelist of the job task information, and the monitoring result is determined based on the monitoring status. The step of determining the task object and task flow nodes based on the task information, and extracting at least two sets of image data to be processed from the monitoring video data according to the task object and the task flow nodes includes: Extract the job number and job equipment information from the job task information, and determine the job object and job process node corresponding to the job number and job equipment information based on the preset job mapping relationship. The preset job mapping relationship includes the mapping relationship between different job numbers, different job equipment information and different job objects and different job process nodes. Determine the monitoring operation node, monitoring time node, and monitoring location node of the work object in the work process node; Based on the monitoring operation node, the monitoring time node, and the monitoring location node, multiple video nodes are selected from the monitoring video data, and at least two sets of image data to be processed are extracted according to the video nodes. The step of filtering multiple video nodes from the monitoring video data based on the monitoring operation node, the monitoring time node, and the monitoring location node includes: Based on the monitoring time node, search all the first reference video nodes, and retrieve the second reference video node corresponding to the target monitoring device from the first reference video nodes according to the monitoring location node; Determine the operation characteristics corresponding to the monitoring operation node, and filter out multiple video nodes from the second reference video nodes according to the operation characteristics; The step of extracting at least two sets of image data to be processed according to the video node includes: Compare the image similarity between multiple image data corresponding to the video node and the pixel range of the operation feature; The comparison results based on the image similarity and the pixel range are used to extract at least two sets of image data to be processed from the image data; The step of extracting target features from the image data according to the monitoring targets in the preset monitoring sequence includes: Retrieve the preset monitoring sequence and query the pointers in the preset monitoring sequence to determine the monitoring target corresponding to the pointers; The image data is processed based on the trained feature recognition model to obtain target features. The feature recognition model is trained on the feature training sample set corresponding to the monitored target.

2. The method according to claim 1, characterized in that, The process of matching the target features with the monitoring whitelist of the task information to obtain the monitoring status of the monitoring video data includes: Obtain the monitoring whitelist of the job task information, the monitoring whitelist including pre-entered security features; When the feature similarity between the target feature and the security feature is greater than a preset threshold, the monitoring status of the surveillance video data is determined to be normal monitoring status. If the feature similarity between the target feature and the security feature is less than or equal to a preset threshold, then the monitoring status of the surveillance video data is determined to be an abnormal monitoring status.

3. The method according to claim 2, characterized in that, The method further includes: Count the number of normal or abnormal monitoring states of the monitoring video data according to the monitoring task information corresponding to the job task information; If the number of states exceeds a preset threshold, the monitoring result is determined to be a monitoring warning, and a warning message is sent. If the number of states does not exceed the preset threshold, the monitoring result is determined to be continuous, and the monitoring video data is updated to re-execute the steps of determining the work object and work process node based on the work task information, and extracting at least two sets of image data to be processed from the monitoring video data according to the work object and the work process node.

4. A system for processing surveillance video data, characterized in that, include: The acquisition module is used to acquire the job task information of each monitored object and the monitoring video data matched with the monitored object; The interception module is used to determine the task object and task process node based on the task information, and to intercept at least two sets of image data to be processed from the monitoring video data according to the task object and the task process node. The extraction module is used to extract the image data according to the monitoring targets in the preset monitoring sequence to obtain target features. The pointer in the preset monitoring sequence is to poll different monitoring targets in time sequence. The matching module is used to match the target features with the monitoring whitelist of the job task information to obtain the monitoring status of the monitoring video data, and to determine the monitoring result based on the monitoring status. The interception module is specifically used to extract the job number and job equipment information from the job task information, and determine the job object and job process node corresponding to the job number and job equipment information based on the preset job mapping relationship. The preset job mapping relationship includes the mapping relationship between different job numbers, different job equipment information and different job objects and different job process nodes. Determine the monitoring operation node, monitoring time node, and monitoring location node of the work object in the work process node; Based on the monitoring operation node, the monitoring time node, and the monitoring location node, multiple video nodes are selected from the monitoring video data, and at least two sets of image data to be processed are extracted according to the video nodes. The interception module is further configured to search all first reference video nodes based on the monitoring time node, and retrieve the second reference video node corresponding to the target monitoring device from the first reference video nodes according to the monitoring location node. Determine the operation characteristics corresponding to the monitoring operation node, and filter out multiple video nodes from the second reference video nodes according to the operation characteristics; The cropping module is further configured to compare the image similarity between multiple image data corresponding to the video node and the pixel range of the operation feature; and to crop at least two sets of image data to be processed from the image data based on the comparison results of the image similarity and the pixel range. The extraction module is also used to retrieve the preset monitoring sequence, query the pointer in the preset monitoring sequence, and determine the monitoring target corresponding to the pointer; The image data is processed based on the trained feature recognition model to obtain target features. The feature recognition model is trained on the feature training sample set corresponding to the monitored target.

5. A storage medium storing at least one executable instruction that causes a processor to perform an operation corresponding to the method for processing surveillance video data as described in any one of claims 1-3.

6. A terminal, comprising: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction, which causes the processor to perform the operation corresponding to the method for processing surveillance video data as described in any one of claims 1-3.

Images