Video data processing method and device, storage medium and electronic equipment

Abstract

The invention discloses a video data processing method and device, a storage medium and electronic equipment. The method comprises the steps that shooting configuration information corresponding to target shooting equipment is acquired, the shooting configuration information comprises a first analysis rule corresponding to a preset point and a second analysis rule corresponding to a non-preset point, the first analysis rule comprises scene features, and the second analysis rule does not comprise the scene features; in response to the target shooting equipment entering a preset point for shooting, generating a first detection result based on the first analysis rule and a first scene feature preset for the preset point; and in response to the target shooting equipment entering a non-preset point for shooting, identifying a second scene feature shot by the target shooting equipment at the non-preset point, and generating a second detection result based on a second analysis rule and the second scene feature. The technical problem of low analysis accuracy of the video data due to single video data processing strategy is solved.

Description

Technical Field

[0001] This application relates to the field of computers, and in particular, to a method and device for processing video data, a storage medium, and an electronic device. Background Art

[0002] Currently, in the process of multi-directionally collecting videos by video shooting devices (such as cloud PTZ cameras, focus zoom PTZ cameras, etc.), existing video data processing strategies often adopt the method of configuring preset points of video shooting devices in a single way, that is, each preset point is configured with fixed intelligent analysis rules, such as vehicle recognition or pedestrian detection. Obviously, the prior art does not consider how to collect videos in the corresponding directions when not at the preset points or when the position and angle of the PTZ camera change. This single rule configuration often cannot adapt to the changing environment, resulting in a large number of false alarms and reducing the overall accuracy of video processing. In summary, there is a technical problem in the related art that the analysis accuracy rate of video data is low due to the single video data processing strategy.

[0003] For the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] Embodiments of this application provide a method and device for processing video data, a storage medium, and an electronic device, so as to at least solve the technical problem that the analysis accuracy rate of video data is low due to the single video data processing strategy.

[0005] According to one aspect of the embodiments of this application, a method for processing video data is provided, including: obtaining shooting configuration information corresponding to a target shooting device, where shooting points of the target shooting device include preset points and non-preset points, the shooting configuration information includes a first analysis rule corresponding to the preset point and a second analysis rule corresponding to the non-preset point, the first analysis rule includes scene features, and the second analysis rule does not include the scene features; in response to the target shooting device entering the preset point for shooting, generating a first detection result based on the first analysis rule and the first scene features preset for the preset point; in response to the target shooting device entering the non-preset point for shooting, identifying second scene features captured by the target shooting device at the non-preset point, and generating a second detection result based on the second analysis rule and the second scene features.

[0006] According to another aspect of the embodiments of the present application, there is also provided a video data processing device, including: an acquisition module, configured to acquire shooting configuration information corresponding to a target shooting device, where the shooting points of the target shooting device include preset points and non-preset points, the shooting configuration information includes a first analysis rule corresponding to the preset points and a second analysis rule corresponding to the non-preset points, the first analysis rule includes scene features, and the second analysis rule does not include the scene features; a first generation module, configured to, in response to the target shooting device entering the preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point; a second generation module, configured to, in response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features.

[0007] Optionally, the device is configured to, in response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features in the following manner: in response to the target shooting device entering the non-preset point for shooting, perform scene feature recognition on the acquisition area of the target shooting device by using a preset recognition method to determine the second scene features; determine the second analysis rule based on the second scene features, and perform target detection on the acquisition area of the target shooting device based on the second analysis rule to determine the second detection result.

[0008] Optionally, the device is configured to, in response to the target shooting device entering the non-preset point for shooting, perform scene feature recognition on the acquisition area of the target shooting device by using a preset recognition method to determine the second scene features in the following manner: in response to the target shooting device entering the non-preset point for shooting, perform recognition on the acquisition area by using a first recognition method to determine a target demarcation line; perform recognition on the acquisition area by using a second recognition method to determine at least one area to be shot; perform scene feature recognition on the target shooting area by using the preset recognition method to determine the second scene features, where the target shooting area represents an area determined based on the target demarcation line and the at least one area to be shot.

[0009] Optionally, the device is configured to respond to the target shooting device entering the preset point for shooting, and generate a first detection result based on the first analysis rule and the first scene feature preset for the preset point, including: obtaining a target identifier corresponding to the preset point entered by the target shooting device; determining the first analysis rule and the first scene feature based on the target identifier; generating a first detection result based on the first analysis rule and the first scene feature.

[0010] Optionally, the device is further configured to: obtain a quality identifier corresponding to the first detection result or the second detection result; in the case where the quality identifier is a first quality identifier, determine a target analysis result based on the first detection result or the second detection result; in the case where the quality identifier is a second quality identifier, determine a target analysis result based on the first detection result or the second detection result, and perform early warning reporting, where the quality level corresponding to the first quality identifier is higher than the quality level corresponding to the second quality identifier; in the case where the quality identifier is a third quality identifier, discard the corresponding detection result and perform early warning reporting, where the quality level corresponding to the second quality identifier is higher than the quality level corresponding to the third quality identifier.

[0011] Optionally, the device is further configured to: in response to a change in the shooting point, perform scene recognition on the current shooting point corresponding to the target shooting device to obtain a target scene result, where the target scene result is used to determine the acquisition area of the target shooting device; adaptively update the first analysis rule or the second analysis rule based on the target scene result.

[0012] Optionally, the device is configured to adaptively update the first analysis rule or the second analysis rule based on the target scene result in the following manner: in the case where the target scene result represents a long-distance scene, set the first analysis rule or the second analysis rule to perform target detection on a first range, and set the detection threshold of the target detection to a first value, and update the first analysis rule or the second analysis rule; in the case where the target scene result represents a medium-distance scene, set the first analysis rule or the second analysis rule to perform target detection on a second range, and set the detection threshold of the target detection to a second value, and update the first analysis rule or the second analysis rule, where the second range is smaller than the first range, and the second value is smaller than the first value; in the case where the target scene result represents a close-distance scene, set the first analysis rule or the second analysis rule to perform target detection on a third range, and set the detection threshold of the target detection to a third value, and update the first analysis rule or the second analysis rule, where the third range is smaller than the second range, and the third value is smaller than the second value.

[0013] According to another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the above-mentioned video data processing method when running.

[0014] According to another aspect of the embodiments of the present application, there is provided a computer program product or a computer program. The computer program product or the computer program includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the video data processing method as described above.

[0015] According to another aspect of the embodiments of the present application, there is also provided an electronic device including a memory and a processor. A computer program is stored in the memory, and the processor is configured to execute the above-mentioned video data processing method through the computer program.

[0016] In the embodiments of the present application, by manually configuring a first analysis rule and first scene features (such as lane lines and detection areas) at a preset point, the accurate matching of the analysis rule and the monitoring scene is ensured, and false alarms are reduced. Specifically, second scene features can be extracted through automatic scene recognition at non-preset points, and combined with a second analysis rule set by default, a detection rule applicable to the current scene is automatically generated to adapt to changing scenes, completing the intelligent analysis of video data, and solving the technical problem of low analysis accuracy of video data. At the same time, due to the introduction of a video quality judgment mechanism, when the video quality is low, automatic analysis stop can be achieved, avoiding analysis errors caused by poor video quality, further improving the overall performance of analyzing video data, effectively combining target recognition and scene classification, dynamically adjusting rule parameters according to different scene features, achieving the purpose of effectively improving the accuracy and reliability of video data analysis. Thus, the technical effect of full-scene intelligent analysis of video data is realized, and further the technical problem of low analysis accuracy of video data due to a single video data processing strategy is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments and descriptions thereof of the present application are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0018] Figure 1 is a schematic diagram of an application environment of an optional video data processing method according to the embodiments of the present application;

[0019] Figure 2It is a schematic flowchart of an optional method for processing video data according to an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of an optional preset point for processing video data according to an embodiment of the present application;

[0021] Figure 4 It is a schematic flowchart of an optional method for processing video data according to an embodiment of the present application;

[0022] Figure 5 It is a schematic diagram of a scenario rule for an optional method for processing video data according to an embodiment of the present application;

[0023] Figure 6 It is a schematic structural diagram of an optional device for processing video data according to an embodiment of the present application;

[0024] Figure 7 It is a schematic structural diagram of an optional product for processing video data according to an embodiment of the present application;

[0025] Figure 8 It is a schematic structural diagram of an optional electronic device according to an embodiment of the present application. Detailed implementation manners

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data used can be interchanged under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] The present application will be described below with reference to the embodiments:

[0029] According to one aspect of the embodiments of the present application, a method for processing video data is provided. Optionally, in this embodiment, the above method for processing video data may be applied to, for example, Figure 1 the hardware environment composed of a server 101 and a terminal device 103 as shown. As Figure 1 shown, the server 101 is connected to the terminal device 103 through a network and can be used to provide services for the terminal device or the application installed on the terminal device. The application 107 can be a video application, an instant messaging application, a browser application, an educational application, a game application, etc. A database 105 can be set up on the server or independently of the server to provide data storage services for the server 101. For example, a game data storage server. The above network may include, but is not limited to: a wired network, a wireless network. Among them, the wired network includes: a local area network, a metropolitan area network, and a wide area network. The wireless network includes: Bluetooth, WIFI, and other networks that implement wireless communication. The terminal device 103 can be a terminal configured with an application and may include, but is not limited to, at least one of the following: a mobile phone (such as an Android mobile phone, an iOS mobile phone, etc.), a laptop computer, a tablet computer, a handheld computer, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, a smart voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a mixed reality (MR) terminal, and other computer devices. The above server can be a single server, a server cluster composed of multiple servers, or a cloud server.

[0030] Combined with Figure 1 shown, the above method for processing video data can be executed by an electronic device. The electronic device can be a terminal device or a server. The above method for processing video data can be implemented separately by the terminal device or the server, or jointly implemented by the terminal device and the server.

[0031] The above is only an example, and this embodiment does not make specific limitations.

[0032] Optionally, as an alternative implementation, as Figure 2 shown, the above method for processing video data includes:

[0033] S202. Obtain the shooting configuration information corresponding to the target shooting device. The shooting points of the target shooting device include preset points and non-preset points. The shooting configuration information includes a first analysis rule corresponding to the preset point and a second analysis rule corresponding to the non-preset point. The first analysis rule includes scene features, and the second analysis rule does not include scene features.

[0034] Optionally, in the embodiments of the present application, the above-mentioned target shooting device refers to an image and video acquisition device with rotation, tilting, and zoom functions, including but not limited to monitoring cameras, intelligent dome cameras, etc., which can provide multi-directional video acquisition services in multiple application scenarios, including but not limited to being deployed in various monitoring scenarios such as traffic monitoring and public security to provide comprehensive video acquisition and intelligent analysis services. The above-mentioned preset point refers to a point with a specific monitoring orientation and angle preset by the target shooting device. At the above-mentioned preset point, the target shooting device can quickly locate and perform video acquisition. The above-mentioned non-preset point refers to any other position except the preset point. The monitoring orientation and angle of the target shooting device at these positions are not preset in advance and may be in a state of random rotation or tilting. The above-mentioned scene features are related to the deployment scenario of the target shooting device and include but are not limited to object features, environmental features, background features, etc. in the video captured by the target shooting device.

[0035] Exemplarily, in the application scenario of traffic monitoring, taking the above-mentioned target shooting device as a PTZ dome camera as an example, when the PTZ dome camera is at the preset point position, it will automatically read the first analysis rule and scene features corresponding to the preset point, such as the drawn lane lines and pedestrian detection areas. At this time, the PTZ dome camera can use these pre-configured rules and features to analyze the video data to ensure the best analysis effect and the lowest false alarm rate in the known monitoring scenario.

[0036] It should be noted that the embodiments of the present application consider some video acquisition areas without preset points. By establishing a second analysis rule corresponding to the non-preset point, scene features can be dynamically generated according to real-time scene recognition. Real-time scene recognition includes but is not limited to deep learning algorithms such as convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), etc.; object detection algorithms such as YOLO, Faster R-CNN, SSD, etc., which are used to automatically identify and locate specific targets in the video; image segmentation algorithms such as fully convolutional network (FCN), U-Net, Mask R-CNN, etc., which are used to segment the video frame into different regions; feature extraction algorithms such as scale-invariant feature transform (SIFT), speeded-up robust features (SURF), oriented FAST and rotated BRIEF (ORB), etc.

[0037] Exemplarily, when the target shooting device is not at the preset point position, it will automatically identify features such as lane lines and pedestrian activity areas in the current scene, and combine default second analysis rules, such as vehicle recognition parameters and pedestrian detection thresholds, to generate an analysis rule applicable to the current non-preset point position, ensuring the accuracy and effectiveness of the analysis results.

[0038] In an exemplary embodiment, Figure 3 is a schematic diagram of a processing preset point for optional video data according to an embodiment of the present application, taking a traffic detection scenario as an example, such as Figure 3 shown as follows:

[0039] When the target shooting device is at preset point 1 and preset point 2, manually configure the above first rules for the positions of preset point 1 and preset point 2, and manually draw the detection area and lane lines according to the camera images of the current preset points; when not at the preset points, the user only needs to configure the default above-mentioned second detection rules, and the lane line recognition and area detection are automatically completed. Through the method of manual configuration at the preset points and automatic recognition at non-preset points, thus, accurate recognition can be achieved at the preset point positions, and at non-preset points, with the help of the scene segmentation module and lane line recognition module inside the real-time scene recognition algorithm, automatic area recognition and lane line recognition are performed, realizing the function of the available algorithm effect, and finally achieving the purpose of full-scene analysis for both preset points and non-preset points.

[0040] Furthermore, at different non-preset point positions, the automatically recognized scene features and generated rules may be different, effectively improving the flexibility and adaptability of video shooting and processing, being able to cope with complex and changeable actual monitoring environments, and avoiding the technical problem of low analysis accuracy rate caused by a single video data processing strategy.

[0041] S204, in response to the target shooting device entering the preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point;

[0042] Optionally, in the embodiment of the present application, the above-mentioned first scene features are related to the deployment scenario of the target shooting device, including but not limited to object features, environmental features, background features, etc. in the video captured by the target shooting device. Taking the application scenario of traffic monitoring as an example, the first scene features include but not limited to lane lines, road areas, fences, traffic lights, etc.

[0043] Exemplarily, in the application scenario of traffic monitoring, when the target shooting device receives the control instruction of "enter preset point 1", that is, in response to the target shooting device entering the preset point for shooting, it automatically calls the first analysis rules pre-set for preset point 1, such as vehicle detection and pedestrian recognition, as well as the first scene features corresponding to preset point 1, such as the lane lines and pedestrian passage areas manually drawn by touch screen or other means on the shooting screen displayed on the intelligent analysis platform corresponding to the target shooting device. The target shooting device will perform intelligent analysis on the collected video data based on the first analysis rules to generate the first detection results. For example, it counts the number of vehicles passing through the lane lines of preset point 1 and identifies abnormal behaviors in the pedestrian passage area, etc.

[0044] It should be noted that when shooting at the preset point, the target shooting device can also be connected to different intelligent analysis platforms. There may be differences in the setting and calling methods of the first analysis rules and the first scene features for each platform. In addition, the types and quantities of preset points also vary according to actual monitoring requirements. For example, in some monitoring scenarios, it may be necessary to distinguish the shooting rules for preset points during the day and at night, and in some scenarios, more preset points may be required to achieve 360-degree panoramic monitoring.

[0045] It should also be noted that under different preset points, the setting of the first scene features (such as the shape and position of lane lines and monitoring areas) can also be adjusted according to the actual scene to adapt to various monitoring environments and requirements.

[0046] S206, in response to the target shooting device entering non-preset point shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate the second detection result based on the second analysis rules and the second scene features.

[0047] Optionally, in the embodiments of the present application, the above non-preset point shooting refers to the monitoring state where the above target shooting device is not at the previously defined preset point position but is in an arbitrary position under free rotation or manual control; the above second scene features refer to the monitoring area features automatically obtained by the target shooting device through real-time scene recognition technology at the non-preset point position, including but not limited to lane lines, pedestrian activity areas, obstacles, etc.

[0048] Exemplarily, in the application scenario of public security, assume that the target shooting device is in a non-preset point position for monitoring. At this time, there are no pre-drawn specific scene features in the monitoring area; the target shooting device will start the real-time scene recognition algorithm to automatically identify second scene features such as pedestrian activity areas and potential gathering points in the monitoring screen. Then, based on the second scene features, the target shooting device will combine the second analysis rules, such as pedestrian density monitoring algorithms and abnormal behavior recognition algorithms, etc., to perform intelligent analysis on the current video data and generate the second detection results, such as crowd density analysis and abnormal behavior early warning, etc.

[0049] It should be noted that during the non-preset point shooting process, when the above target shooting device performs second scene feature recognition, it may encounter different types of environments, such as indoor, outdoor, day, night, weather changes, etc. Therefore, the above second analysis rule in the embodiments of the present application may also need to be adaptively adjusted according to scene changes. For example, when monitoring at night, the lighting conditions are taken into account and the parameters of the image processing algorithm are adjusted; when monitoring large public events, the sensitivity of pedestrian density monitoring is increased to meet more complex monitoring requirements.

[0050] In an exemplary embodiment, Figure 4 is a schematic flowchart of an optional video data processing method according to the embodiments of the present application. Taking the traffic monitoring scene as an example, as Figure 4 shown:

[0051] S402, start;

[0052] S404, obtain the preset point information of the target shooting device;

[0053] S406, configure the preset point rule (the above first analysis rule);

[0054] S408, configure the default rule (the above second analysis rule);

[0055] S410, determine whether the target shooting device is at the preset point position. If so, execute S412; otherwise, execute S416;

[0056] S412, obtain the preset point number of the current target shooting device;

[0057] S414, obtain the intelligent analysis rule set for the preset point, and then execute S420;

[0058] S416, perform automatic area recognition and lane line recognition;

[0059] S418, obtain the default rule configuration;

[0060] S420, obtain the current scene, including the long-shot scene, medium-shot scene, and close-up scene;

[0061] S422, obtain the scene parameter configuration corresponding to the rule;

[0062] S424, send the combined intelligent analysis rule to the target shooting device;

[0063] S426, judge the video quality of the current channel to obtain the video quality diagnosis result;

[0064] S428-1, the video quality diagnosis result indicates that the video shooting quality is high, and the intelligent analysis of video data continues;

[0065] S428-2, the video quality diagnosis result indicates that the video shooting quality is medium, and early warning analysis is carried out (the video quality does not meet the expectation, which will affect the accuracy of intelligent analysis);

[0066] S428-3, the video quality diagnosis result indicates that the video shooting quality is medium, and the analysis is stopped;

[0067] S420, end.

[0068] Specifically, Figure 5 is a schematic diagram of the scenario rules of an optional video data processing method according to an embodiment of the present application. As Figure 5 shown, among them, the intrusion of the mixing line in road detection refers to a security protection measure used to prevent unauthorized personnel or vehicles from entering a specific area or road. The mixing line is a physical obstacle, usually made of metal wire, plastic wire or other materials, and can be visible or concealed; when someone or a vehicle touches these mixing lines, the alarm system will be triggered to notify relevant personnel or activate other security measures. The above long-range scenarios include, but are not limited to, large-scale monitoring areas with relatively small targets. At this time, the threshold of target detection in intelligent analysis can be appropriately increased to reduce the occurrence of false alarms; the medium-range scenarios include, but are not limited to, scenarios with a moderate target size at a monitoring distance, and appropriate target detection thresholds and sensitivities can be set to ensure better analysis effects; the above close-range scenarios include, but are not limited to, small-scale monitoring with relatively large targets, and the threshold and sensitivity of target detection can be reduced to ensure the integrity and high accuracy of the targets. When the preset point of the target shooting device rotates, the scene is recognized, and the rule parameters corresponding to the scene are matched. By combining target recognition with scene classification and setting corresponding rule parameters according to the characteristics of the targets in different scenarios, accurate discrimination of long-range, medium-range and close-range scenarios and adaptive adjustment of rule parameters can be achieved, thereby effectively improving the accuracy and reliability of intelligent analysis in traffic monitoring.

[0069] It should also be noted that in the embodiment of the present application, the above first detection result and second detection result can be associated. In other words, a more detailed and specific full-scene intelligent analysis can be obtained by combining the first detection result of the non-preset point and the first detection result of the preset point. For example:

[0070] Taking the intelligent monitoring system of a shopping mall as an example, when the target shooting device (such as a PTZ camera) enters a preset point (such as the entrance, escalator, cashier desk, etc.) for shooting, the system will call the first analysis rule corresponding to the preset point, and combine it with the first scene features (such as the pedestrian flow monitoring area at the entrance, the safety monitoring line of the escalator, the queuing area in front of the cashier desk, etc.) to generate the first detection result, such as early warning of abnormal personnel gathering, detection of safety violation behaviors, etc.; when the camera enters a non-preset point for shooting, for example, when freely rotating to monitor the stores in the mall, the system will start the real-time scene recognition technology, automatically identify the customer activity area in front of the store (the second scene feature), and combine it with the second analysis rule (such as customer behavior analysis, abnormal event monitoring, etc.) to generate the second detection result, such as customer interest point analysis, early warning of abnormal behaviors, etc.; the first detection result and the second detection result can be further combined for intelligent analysis of the entire scene. For example, by comparing the customer flow and behavior patterns at different preset points and non-preset points, a more comprehensive customer activity analysis report can be obtained, providing data support for the management and operation of the shopping mall.

[0071] Through the embodiments of the present application, by manually configuring the first analysis rule and the first scene features (such as lane lines, detection areas) at the preset point, the accurate matching of the analysis rule and the monitoring scene is ensured, reducing false alarms; specifically, at non-preset points, the second scene features can be extracted through automatic scene recognition, combined with the default second analysis rule, and the detection rules applicable to the current scene are automatically generated to adapt to the changing scene, completing the intelligent analysis of video data, achieving the purpose of effectively improving the accuracy and reliability of video data analysis, thus realizing the technical effect of intelligent analysis of the entire scene of video data, and further solving the technical problem of low analysis accuracy of video data.

[0072] As an optional solution, in response to the target shooting device entering the non-preset point for shooting, identifying the second scene features captured by the target shooting device at the non-preset point, and generating the second detection result based on the second analysis rule and the second scene features, including: in response to the target shooting device entering the non-preset point for shooting, using a preset recognition method to perform scene feature recognition on the acquisition area of the target shooting device to determine the second scene features; determining the second analysis rule based on the second scene features, and performing target detection on the acquisition area of the target shooting device based on the second analysis rule to determine the second detection result.

[0073] Optionally, in the embodiments of the present application, the above acquisition area refers to the lane lines and the road area in the shooting scene; the above preset recognition method refers to the algorithms built in the intelligent analysis system of the target shooting device for automatically identifying the targets and scenes in the current monitoring screen, including but not limited to deep learning object detection, image segmentation technology, motion detection, etc.

[0074] Exemplarily, in the application scenario of large public event monitoring, when the target shooting device (such as a PTZ camera) enters non-preset point shooting, that is, when the monitoring personnel do not set a specific preset point to monitor the activity area, the intelligent analysis system of the camera will automatically start the preset recognition method to perform real-time recognition on the current area, determine the acquisition area, and obtain the lane lines and the road area.

[0075] Furthermore, based on the above second scenario feature, the target shooting device dynamically determines the second analysis rule. For example, higher face recognition and abnormal behavior detection sensitivity are set for crowded areas. Finally, target detection is performed on the acquisition area based on the second analysis rule to determine the second detection result, such as abnormal behavior warning, etc.

[0076] In an exemplary embodiment, taking the target shooting device as a PTZ camera installed in a shopping mall as an example, when the camera enters non-preset point shooting, the intelligent analysis system identifies the acquisition area in the current monitoring screen through preset image segmentation and target detection algorithms. At this time, the acquisition area can be the customer queuing area in front of the vending machine, the children's play area, etc. Then, according to the identified second scenario feature, the second analysis rule is dynamically generated and adjusted. For example, the crowd flow monitoring threshold for the customer queuing area is set, and the motion detection algorithm parameters for the children's play area are optimized. Target detection is performed on the target shooting area based on the adjusted second analysis rule to determine the second detection results such as warning for overly long customer queue and warning for abnormal behavior detection in the play area, improving the service efficiency and safety of the shopping mall.

[0077] Through the embodiments of the present application, by adopting the dynamic scenario recognition and intelligent rule adaptive adjustment technology, it is realized to automatically identify the key monitoring areas (second scenario features) during non-preset point shooting, and dynamically generate the second analysis rule according to these features for accurate target detection and intelligent analysis, achieving the purpose of maintaining high analysis accuracy and analysis efficiency of video data in various non-preset point monitoring scenarios.

[0078] As an alternative solution, the above-mentioned response to the target shooting device entering the non-preset point shooting, adopting the preset recognition method to perform scenario feature recognition on the acquisition area of the target shooting device to determine the second scenario feature, includes: in response to the target shooting device entering the non-preset point shooting, using the first recognition method to recognize the acquisition area to determine the target demarcation line; using the second recognition method to recognize the acquisition area to determine at least one area to be shot; using the preset recognition method to perform scenario feature recognition on the target shooting area to determine the second scenario feature, where the target shooting area represents the area determined based on the target demarcation line and the at least one area to be shot.

[0079] Optionally, in the embodiments of the present application, the above-mentioned target demarcation line may be a lane line as shown in Figure 3 , and the above-mentioned area to be photographed may be a detection area as shown in Figure 3 .

[0080] In an exemplary embodiment, in the application scenario of highway monitoring, when the above-mentioned target photographing device (such as a PTZ camera) enters non-preset point photographing, a first recognition method is started, such as a lane line recognition algorithm, to automatically analyze the lane distribution in the current monitoring screen and determine the target demarcation line, that is, the demarcation lines of each lane, as shown in Figure 3 . The target demarcation line includes but is not limited to differentiating driving lanes, and may also include an emergency lane, an access lane, etc.

[0081] In yet another exemplary embodiment, when the above-mentioned target photographing device (such as a PTZ camera) enters non-preset point photographing, a second recognition method may also be started, such as a region detection algorithm, to further analyze the monitoring screen and identify at least one area to be photographed, such as identifying a crosswalk area (detection area) near the exit, or an emergency stop area at the edge of the monitoring screen, etc.

[0082] Furthermore, based on the above-mentioned determined target demarcation line (lane line) and the above-mentioned at least one area to be photographed (detection area), the target photographing area may be comprehensively judged and determined. For example, all lanes and the areas to be photographed (such as crosswalks, emergency stop areas) around them are combined into a continuous target photographing area to ensure that the key areas of the entire highway monitoring screen are covered, so as to perform intelligent analysis of the entire scene, such as vehicle violation detection, pedestrian intrusion warning, emergency event discovery, etc.

[0083] Through the embodiments of the present application, the first recognition method and the second recognition method are used to perform multi-level analysis on the monitoring screen under non-preset points, determine the target demarcation line and the area to be photographed, and then accurately locate the target photographing area, realizing efficient and accurate video intelligent analysis even at non-preset monitoring positions.

[0084] As an alternative solution, the above-mentioned generating a first detection result based on the first analysis rule and the first scene feature preset for the preset point in response to the above-mentioned target photographing device entering the preset point for photographing includes: obtaining the target identifier corresponding to the preset point entered by the above-mentioned target photographing device; determining the first analysis rule and the first scene feature based on the target identifier; generating a first detection result based on the first analysis rule and the first scene feature.

[0085] Optionally, in the embodiments of the present application, the above-mentioned target identifier is used to distinguish each preset point of the target shooting device, and the target identifiers of different preset points are different.

[0086] It should be noted that although the target identifiers of different preset points are different, the first analysis rules and the first scene features of different preset points may be the same or different, which depends on the monitoring requirements between the preset points and the similarity or difference of the scenes. For example, taking the public travel safety monitoring scene as an example, the preset points of multiple exit channels may share a set of first analysis rules for pedestrian density monitoring, while the preset points in the baggage claim area and the security check area may focus on the rules for detecting left items and identifying abnormal behaviors of personnel.

[0087] Exemplarily, in the security monitoring scene of an airport terminal, when the target shooting device (such as a PTZ dome camera) enters a preset point for shooting, first obtain the target identifier corresponding to the preset point entered by the above-mentioned target shooting device. This target identifier can be the unique number or name of the preset point, which is used to distinguish the specific location where the current device is located among multiple preset points, such as "Preset Point No. 1 in the Baggage Claim Area" or "Preset Point No. 3 at the Security Checkpoint". The target identifier of each preset point is unique, ensuring that the monitoring position of the dome camera can be accurately identified.

[0088] Furthermore, based on the obtained target identifier, determine the above-mentioned first analysis rule and the above-mentioned first scene feature. For example, for "Preset Point No. 1 in the Baggage Claim Area", call the first analysis rule, including but not limited to pedestrian density monitoring, detection of suspicious left items, identification of people lingering, etc., and the corresponding first scene features, such as the marked baggage claim belt area, exit channels, rest areas, etc. Then, generate a first detection result based on the above-mentioned first analysis rule and the above-mentioned first scene feature. For example, in "Preset Point No. 1 in the Baggage Claim Area", according to the first analysis rule combined with the first scene features such as the known baggage claim belt area and exit channels, perform real-time video intelligent analysis to generate a first detection result, such as a warning that the pedestrians near the baggage claim belt are too dense, an alarm that there are suspicious left items in the exit channel, etc., so as to assist security personnel in timely discovering and handling potential security problems.

[0089] Through the embodiments of the present application, by implementing steps of obtaining the target identifier, calling the first analysis rule and the first scene feature based on the target identifier, and generating the first detection result, accurate monitoring and intelligent analysis of the preset point are completed. It not only provides customized intelligent analysis services according to the specific monitoring requirements of the preset point, but also can improve the overall resource utilization efficiency of the system by sharing the analysis rules and scene features of similar preset points.

[0090] As an alternative solution, the above method further includes: obtaining a quality identifier corresponding to the first detection result or the second detection result; in the case where the quality identifier is a first quality identifier, determining a target analysis result based on the first detection result or the second detection result; in the case where the quality identifier is a second quality identifier, determining a target analysis result based on the first detection result or the second detection result, and performing early warning reporting, where the quality level corresponding to the first quality identifier is higher than the quality level corresponding to the second quality identifier; in the case where the quality identifier is a third quality identifier, discarding the corresponding detection result and performing early warning reporting, where the quality level corresponding to the second quality identifier is higher than the quality level corresponding to the third quality identifier.

[0091] Optionally, in the embodiments of the present application, the quality identifier refers to an evaluation flag for the credibility or quality of the first detection result or the second detection result, used to distinguish the quality of detection results at different levels. The first quality identifier corresponds to a high-quality detection result, indicating that the first detection result and the second detection result are highly accurate and can intuitively reflect important information in the actual monitoring scenario; the second quality identifier corresponds to a medium-quality detection result, indicating that the first detection result or the second detection result may be affected by environmental factors or algorithm limitations, and its accuracy needs to be further confirmed; the third quality identifier corresponds to a low-quality detection result, indicating that the first detection result or the second detection result has low accuracy and poor video quality.

[0092] Exemplarily, in the application scenario of urban road monitoring, when a target shooting device (such as a PTZ camera) takes pictures at a preset point (such as a main intersection) or a non-preset point (such as a non-main road section) and completes intelligent analysis, a first detection result or a second detection result will be generated, and then the quality identifier corresponding to the first detection result or the second detection result will be obtained to evaluate the credibility of the recognition result.

[0093] For example, when the first detection result or the second detection result indicates that the recognition result of a vehicle violation behavior is very clear, the video picture is clear, and the target features are obvious, a "first quality identifier" will be assigned to this detection result at this time, indicating that it is a high-quality recognition result; in the case where the detection result corresponds to the first quality identifier, directly use this high-quality detection result for subsequent intelligent analysis and data processing, such as counting the frequency of violation behaviors and providing accurate data for traffic management.

[0094] For another example, in the case where the detection result corresponds to the second quality identifier, due to its low accuracy, the target shooting device will perform early warning reporting, including but not limited to sending a warning message to the monitoring center, prompting the monitoring personnel to pay attention to this result and perform manual review or take further monitoring measures to ensure the accuracy of the analysis.

[0095] For another example, in the case where the detection result corresponds to the third quality identifier, since the accuracy of the detection result cannot be guaranteed, the detection result will not be used for intelligent analysis. Instead, a warning message will be sent to indicate possible video quality problems or difficulties in target recognition, so as to avoid making wrong decisions based on inaccurate results.

[0096] Through the embodiments of the present application, by adopting intelligent analysis result evaluation and grading processing based on quality identifiers, precise management and efficient utilization of the first detection result and the second detection result are achieved, and the technical purpose of ensuring the reliability of the analysis result, timely warning of potential problems, and avoiding interference from low-quality data in full-scenario intelligent monitoring is achieved.

[0097] As an optional solution, the above method further includes: in response to a change in the above shooting position, performing scene recognition on the current shooting position corresponding to the above target shooting device to obtain a target scene result, where the above target scene result is used to determine the acquisition area of the above target shooting device; adaptively updating the above first analysis rule or the above second analysis rule based on the above target scene result.

[0098] Optionally, in the embodiments of the present application, the above shooting position refers to the specific position and angle of the target shooting device (such as a PTZ camera) in the monitoring environment, including but not limited to preset points and non-preset points; the above target scene result refers to the scene description result obtained after scene recognition and used to guide the adjustment of the analysis rule, including but not limited to the target type, quantity, distribution, environmental state, etc. within the monitoring area; the above acquisition area refers to the specific area in the monitoring screen determined according to the target scene result and used for subsequent intelligent analysis.

[0099] Exemplarily, in the intelligent monitoring scenario of a large shopping mall, when the shooting position of the target shooting device (such as a PTZ camera) changes, for example, when moving from a preset point (such as the entrance) to a non-preset point (such as near a certain store) for shooting, a comprehensive analysis of the monitoring screen at the current shooting position will be performed to identify information such as the target type (such as customers, store employees), quantity, distribution, and environmental state (such as lighting, crowd density) within the monitoring area, and the above target scene result will be generated. Based on this target scene result, the acquisition area will be determined, and important areas such as customer activities, suspicious behaviors, or product display areas will be focused on.

[0100] Furthermore, if the target shooting device turns from the entrance preset point to a non-preset point of a certain store, the second analysis rule will be automatically updated to adjust the analysis focus to customer behavior patterns, abnormal stay detection, etc. to adapt to the new monitoring requirements; if the target shooting device returns from the non-preset point to the entrance preset point, the first analysis rule can also be automatically restored to refocus on functions such as crowd flow monitoring and entrance security warning.

[0101] It should be noted that considering that the processes of scene recognition and configuration rule updates may be affected by various factors, such as the moving speed of the target shooting device, the complexity of the monitoring environment, the dynamic changes of the target, etc., therefore, the embodiments of the present application also have the capabilities of rapid response and intelligent adjustment to ensure that new scenes can be quickly recognized and analysis strategies can be adjusted after the shooting position changes, so as to maintain the stability and reliability of the monitoring effect.

[0102] In an exemplary embodiment, taking the monitoring of smart agriculture as an example, when the target shooting device (such as an intelligent dome camera installed in a farmland) is monitoring, it will adjust the shooting position at different time periods according to the crop growth stage and farmland management requirements; when the shooting position of the target shooting device changes, the scene recognition algorithm is started to analyze the picture at the current shooting position, and information such as crop growth status, pest and disease distribution, and water source conditions is recognized to generate the target scene result.

[0103] Furthermore, the acquisition area can be adaptively adjusted based on the target scene result, such as setting a higher resolution acquisition in the area with dense crop growth, or increasing the detection frequency in the potentially high-incidence area of pests and diseases. At the same time, the first analysis rule or the second analysis rule can also be updated according to the scene recognition result. For example, during the crop maturity stage, fruit detection and quality assessment rules are added, and during the high-incidence period of pests and diseases, disease warning and pest behavior analysis rules are enabled to meet the farmland monitoring requirements in specific stages.

[0104] Through the embodiments of the present application, by adopting the scene recognition and adaptive rule update technology in response to the change of the shooting position, the dynamic management and efficient utilization of video data are realized, the purpose of optimizing the analysis effect, improving the monitoring pertinence and intelligent level in different monitoring scenes is achieved, and it is ensured that stable and accurate intelligent analysis services can be provided in all scenarios.

[0105] As an alternative solution, adaptively updating the first analysis rule or the second analysis rule based on the above target scenario result includes: when the above target scenario result represents a long-range scenario, setting the first analysis rule or the second analysis rule to perform target detection on a first range, setting the detection threshold of the above target detection to a first value, and updating the first analysis rule or the second analysis rule; when the above target scenario result represents a mid-range scenario, setting the first analysis rule or the second analysis rule to perform target detection on a second range, setting the detection threshold of the above target detection to a second value, and updating the first analysis rule or the second analysis rule, where the second range is smaller than the first range, and the second value is smaller than the first value; when the above target scenario result represents a close-range scenario, setting the first analysis rule or the second analysis rule to perform target detection on a third range, setting the detection threshold of the above target detection to a third value, and updating the first analysis rule or the second analysis rule, where the third range is smaller than the second range, and the third value is smaller than the second value.

[0106] Optionally, in the embodiments of the present application, the above target scenario result refers to the analysis result of the scenario type in the monitoring video obtained according to the scenario recognition technology, including but not limited to long-range scenarios, mid-range scenarios, and close-range scenarios; the first range, the second range, and the third range refer to the effective regions of target detection in the monitoring video adaptively adjusted by the system for different scenario types, and their size relationship is that the close range is smaller than the mid range, and the mid range is smaller than the long range, to adapt to the target characteristics and monitoring requirements at different distances; the first value, the second value, and the third value respectively refer to the detection thresholds of the target detection algorithm under different scenario types, used to control the sensitivity of target detection, and their numerical size relationship is that the first value is smaller than the second value, and the second value is smaller than the third value.

[0107] Exemplarily, in the application scenario of intelligent wildlife protection, when the target scenario result identifies that the current monitoring video is a long-range scenario, it means that the target shooting device (such as a PTZ camera) is located at a relatively long distance to observe the entire protected area or a specific habitat. At this time, the target shooting device sets the first analysis rule or the second analysis rule to perform target detection on a first range, that is, a large range covering the entire protected area, and at the same time sets the detection threshold to a first value to reduce the sensitivity to small targets in the distance, reduce false alarms, and focus on the monitoring of large animals or group behaviors.

[0108] Exemplarily, in the case where a mid-shot scene is recognized, indicating that the target shooting device is monitoring a certain medium-distance area within the protected area, such as a water source area or a specific animal habitat, the target shooting device adjusts the first analysis rule or the second analysis rule to perform target detection on the second range, that is, narrowing the detection range to the periphery of the water source area or the habitat, and at the same time setting the detection threshold to the second value to improve the detection sensitivity for medium-distance targets and ensure more detailed observation and recording of the activities and behaviors of wild animals.

[0109] Exemplarily, when the target scene result indicates that the current picture is a close-up scene and the target shooting device is observing a specific animal or its behavior at close range, such as in an animal medical examination area or a cub feeding area, the target shooting device further optimizes the first analysis rule or the second analysis rule to perform target detection on the third range, only focusing on the animal medical examination area or the cub feeding area, and at the same time setting the detection threshold to the third value to maximize the recognition ability for nearby targets and ensure that every subtle movement or feature can be accurately captured and analyzed.

[0110] Through the embodiments of the present application, by adopting the adaptive target detection configuration technology based on the target scene result, the dynamic adjustment of the target detection range and the detection threshold under different scene types is realized, achieving the purpose of optimizing the target detection effect, reducing false alarms, and improving the detection pertinence and accuracy in the intelligent monitoring system.

[0111] In an exemplary embodiment, the embodiments of the present application can be applied to the field of video analysis and the field of PTZ camera intelligence. Taking the PTZ camera as the target shooting device as an example:

[0112] Considering that in traffic monitoring applications, PTZ cameras usually set multiple preset points to monitor video images in different directions. However, different preset points need to run different intelligent analysis rules, such as vehicle recognition, pedestrian detection, etc.; due to the limitations of the preset points and the changes in the camera image angles, the intelligent analysis area and the lane lines may deviate from the actual scene, resulting in a large number of false alarms.

[0113] In addition, weather reasons or camera occlusion, such as heavy rain and thick fog weather, often lead to a decline in video quality and blurring, thereby reducing the analysis effect and generating a large number of false alarms. Moreover, the rotation and zoom characteristics of the camera may also cause the same set of rule parameters to fail due to changes in the focal length, thereby generating a large number of false alarms.

[0114] To solve the above technical problems, the embodiments of the present application implement an intelligent traffic scene analysis method based on preset points and non-preset points of a PTZ camera. The PTZ camera issues the intelligent rules already configured at the preset points at the preset point positions to ensure the accurate matching of the rules with the analysis positions; at non-preset point positions, the PTZ camera automatically extracts lane lines and regional targets through automatic scene recognition technology and issues intelligent rules to achieve the intelligent analysis function for the entire scene.

[0115] On the other hand, at the preset point positions, the accuracy of the analysis is ensured through manually configured rules and drawn regions and lane lines; at non-preset point positions, scene recognition algorithms are applied to automatically identify regions and lane lines, and the reliability of the analysis for the entire scene is ensured through intelligent rule analysis.

[0116] On yet another hand, in combination with the video quality algorithm, the video quality of the current channel is judged and divided into three grades: high, medium, and low. When the video quality is at the medium grade, a warning prompt is given; when the video quality is at the low grade, the intelligent analysis is stopped to avoid inaccurate results; also, in combination with the target recognition technology, the far view, medium view, and close view are judged, different rule parameters are set, and the rule parameters are adaptively adjusted according to different scenarios to further improve the accuracy of the analysis.

[0117] Exemplarily, the implementation steps of the embodiments of the present application include but are not limited to:

[0118] S1, Preset point configuration: Set multiple preset points within the monitoring area, and each preset point represents a different monitoring orientation. Configure the intelligent configurations to be analyzed (pedestrian detection, parking detection, etc.), regions, lane lines, etc. for each preset point.

[0119] S2, Configure the default intelligent configuration rules of the PTZ camera for the PTZ camera to perform intelligent analysis at non-preset point positions. At this time, regions and lane lines are not configured.

[0120] S3, When the PTZ camera enters the preset point position, according to the configuration information of the preset point, issue the intelligent rules, regions, lane line parameters, etc. already configured at the preset point. At this time, the PTZ camera uses the intelligent configuration at the preset point position for target recognition and intelligent analysis.

[0121] S4, Non-preset point intelligent analysis: Based on the regions and lane lines recognized at non-preset points, the PTZ camera automatically issues the default intelligent rule configuration. Among them, the regions and lane lines can be optimized and adjusted according to real-time region recognition and lane line recognition algorithms to adapt to lane line and region analysis at different angles.

[0122] S5, Video quality judgment: Combine the video quality algorithm to judge the video quality of the current channel, and classify it into three grades: high, medium, and low. Perform corresponding processing according to different video quality grades. Analyze the normal results of high-grade video quality diagnosis. Report early warnings for medium-grade results. Stop intelligent analysis for low-grade results.

[0123] S6, Rule parameter adaptation: Judge the long shot, medium shot, and close shot according to the target recognition technology, set different rule parameters, and perform adaptive adjustment according to different scenarios to improve the accuracy of analysis.

[0124] Through the above steps, the full-scene intelligent analysis function of the PTZ can be realized in traffic monitoring applications, and false alarms can be effectively avoided.

[0125] Specifically, when the PTZ is at the preset point, it is necessary to manually configure the rules for the position of the preset point, and draw the detection area and lane lines manually according to the PTZ image of the current preset point. When not at the preset point, the user only needs to configure the default detection rules, and the lane line recognition and area detection work are automatically completed by the algorithm; through the combination of manual configuration at the preset point and automatic recognition at non-preset points, accurate recognition can be achieved at the preset point position, with good algorithm recognition effects. When not at the preset point, with the help of the scene segmentation module and lane line recognition module inside the algorithm, automatic area recognition and lane line recognition are carried out to achieve the function of available algorithm effects. Finally, the combination of preset points and non-preset points is used to complete the full-scene analysis.

[0126] It should be noted that the rule parameter adaptation scheme includes but is not limited to: generating three sets of templates for the long shot scene, medium shot scene, and close shot scene as shown in Figure 5 The above first analysis rule and second analysis rule include but are not limited to the long shot scene: The long shot usually includes a large-scale monitoring area and relatively small targets; appropriately increase the target detection threshold to reduce the occurrence of false alarms; Medium shot scene: The medium shot is usually a scene at a general monitoring distance, and the target size is moderate; Set an appropriate target detection threshold and sensitivity to ensure a good analysis effect; Close shot scene: The close shot scene is generally a small-scale monitoring, and the target is relatively large; Reduce the target detection threshold and sensitivity to ensure the integrity and high accuracy of the target; Identify the scene when the PTZ rotates at the preset point; and match the rule parameters of the corresponding scene.

[0127] In summary, by combining target recognition with scene classification and setting corresponding rule parameters according to the characteristics of targets in different scenes, accurate discrimination of long-shot, medium-shot, and close-shot scenes and adaptive adjustment of rule parameters can be achieved, thereby improving the accuracy and reliability of intelligent analysis in traffic monitoring. The embodiments of this application implement an intelligent analysis strategy for the entire scene: combining the intelligent rules of preset points configured manually with the automatic scene recognition technology under non-preset points to solve the false alarm problems caused by the limitations of preset points and scene changes; and dynamic scene recognition and lane line and region extraction algorithms: The PTZ can automatically recognize the scene and extract lane lines and monitoring regions at non-preset point positions, which can improve the intelligence level of the PTZ and achieve more accurate and efficient monitoring and analysis functions; the mechanism of dynamic issuance of intelligent rules: adjust the analysis rules in real time according to the position status of the PTZ (preset point or non-preset point), improving the adaptability and accuracy of the system; the video quality judgment method: combine the video quality algorithm to judge the video quality of the current channel and divide it into three grades: high, medium, and low. According to different video quality grades, corresponding processing is carried out. In the case of high-quality video, intelligent analysis is carried out normally. A warning prompt is issued in the case of medium-quality video, and intelligent analysis is stopped in the case of low-quality video to avoid inaccurate results; the parameter adaptive adjustment function: judge long-shot, medium-shot, and close-shot according to the target recognition technology, set different rule parameters, and adaptively adjust the rule parameters according to different scenes to further improve the accuracy of analysis.

[0128] Through the embodiments of this application, full-scene intelligent analysis is realized by using preset point configuration and non-preset point scene recognition; the rule parameters are adaptively adjusted according to the scene to improve accuracy; combined with video quality judgment to avoid false alarms caused by low-quality video; high flexibility and can save human resources; optimize the rule parameters in real time to keep the analysis effect stable.

[0129] It can be understood that in the specific implementation of this application, when it comes to data related to user information, etc., when the above embodiments of this application are applied to specific products or technologies, user permission or consent needs to be obtained, and the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions.

[0130] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0131] According to another aspect of the embodiments of the present application, there is also provided a video data processing device for implementing the above-mentioned video data processing method. As Figure 6 shown, the device includes:

[0132] An acquisition module 602, configured to acquire shooting configuration information corresponding to a target shooting device, where the shooting points of the target shooting device include preset points and non-preset points, the shooting configuration information includes a first analysis rule corresponding to the preset points and a second analysis rule corresponding to the non-preset points, the first analysis rule includes scene features, and the second analysis rule does not include scene features;

[0133] A first generation module 604, configured to, in response to the target shooting device entering a preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point;

[0134] A second generation module 606, configured to, in response to the target shooting device entering a non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features.

[0135] As an alternative solution, the above-mentioned device is configured to, in response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features in the following manner: in response to the target shooting device entering the non-preset point for shooting, perform scene feature recognition on the acquisition area of the target shooting device by using a preset recognition method to determine the second scene features; determine the second analysis rule based on the second scene features, and perform target detection on the acquisition area of the target shooting device based on the second analysis rule to determine the second detection result.

[0136] As an alternative solution, the above-mentioned device is configured to, in response to the target shooting device entering the non-preset point for shooting, perform scene feature recognition on the acquisition area of the target shooting device by using a preset recognition method to determine the second scene features in the following manner: in response to the target shooting device entering the non-preset point for shooting, perform recognition on the acquisition area by using a first recognition method to determine a target demarcation line; perform recognition on the acquisition area by using a second recognition method to determine at least one area to be shot; perform scene feature recognition on the target shooting area by using the preset recognition method to determine the second scene features, where the target shooting area represents an area determined based on the target demarcation line and the at least one area to be shot.

[0137] As an alternative solution, the above-mentioned device is used to respond to the target shooting device entering the preset point for shooting, and generate a first detection result based on the first analysis rule and the first scene feature preset for the preset point, including: obtaining the target identifier corresponding to the preset point entered by the target shooting device; determining the first analysis rule and the first scene feature based on the target identifier; generating a first detection result based on the first analysis rule and the first scene feature.

[0138] As an alternative solution, the above-mentioned device is further used to: obtain the quality identifier corresponding to the first detection result or the second detection result; in the case where the quality identifier is the first quality identifier, determine the target analysis result based on the first detection result or the second detection result; in the case where the quality identifier is the second quality identifier, determine the target analysis result based on the first detection result or the second detection result, and perform early warning reporting, where the quality level corresponding to the first quality identifier is higher than the quality level corresponding to the second quality identifier; in the case where the quality identifier is the third quality identifier, discard the corresponding detection result and perform early warning reporting, where the quality level corresponding to the second quality identifier is higher than the quality level corresponding to the third quality identifier.

[0139] As an alternative solution, the above-mentioned device is further used to: in response to a change in the shooting position, perform scene recognition on the current shooting position corresponding to the target shooting device to obtain a target scene result, where the target scene result is used to determine the acquisition area of the target shooting device; adaptively update the first analysis rule or the second analysis rule based on the target scene result.

[0140] As an alternative solution, the above-mentioned device is used to adaptively update the first analysis rule or the second analysis rule based on the target scene result in the following manner: in the case where the target scene result represents a long-shot scene, set the first analysis rule or the second analysis rule to perform target detection on the first range, and set the detection threshold of the target detection to the first value, and update the first analysis rule or the second analysis rule; in the case where the target scene result represents a medium-shot scene, set the first analysis rule or the second analysis rule to perform target detection on the second range, and set the detection threshold of the target detection to the second value, and update the first analysis rule or the second analysis rule, where the second range is smaller than the first range and the second value is smaller than the first value; in the case where the target scene result represents a close-shot scene, set the first analysis rule or the second analysis rule to perform target detection on the third range, and set the detection threshold of the target detection to the third value, and update the first analysis rule or the second analysis rule, where the third range is smaller than the second range and the third value is smaller than the second value.

[0141] In the embodiments of the present application, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, which works together with other relevant parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the function of the module or unit.

[0142] Regarding the device in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0143] According to one aspect of the present application, a computer program product is provided, and the computer program product includes a computer program.

[0144] The serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0145] Figure 7 A block diagram of a computer system of an electronic device for implementing the embodiments of the present application is schematically shown.

[0146] It should be noted that Figure 7 The computer system 700 of the electronic device shown is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present application.

[0147] As Figure 7 shown, the computer system 700 includes a central processing unit 701 (Central Processing Unit, CPU), which can perform various appropriate actions and processes according to a program stored in a read-only memory 702 (Read-Only Memory, ROM) or a program loaded from a storage section 708 into a random access memory 703 (Random Access Memory, RAM). In the random access memory 703, various programs and data required for system operation are also stored. The central processing unit 701, the read-only memory 702, and the random access memory 703 are connected to each other through a bus 704. An input / output interface 705 (Input / Output interface, i.e., I / O interface) is also connected to the bus 704.

[0148] The following components are connected to the input / output interface 705: an input part 706 including a keyboard, a mouse, etc.; an output part 707 including such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage part 708 including a hard disk, etc.; and a communication part 709 including a network interface card such as a local area network card, a modem, etc. The communication part 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output interface 705 as required. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as required so that a computer program read from it can be installed into the storage part 708 as required.

[0149] Specifically, according to an embodiment of the present application, the processes described in each method flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit 701, various functions defined in the system of the present application are executed.

[0150] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit 701, various functions provided by the embodiments of the present application are executed.

[0151] According to another aspect of the embodiments of the present application, an electronic device for implementing the management method of the above system is further provided. The electronic device can be Figure 1 the terminal device or server shown. This embodiment takes the electronic device as the terminal device as an example for illustration. As Figure 8 shown, the electronic device includes a memory 802 and a processor 804. A computer program is stored in the memory 802, and the processor 804 is configured to execute the steps in any one of the above method embodiments through the computer program.

[0152] Optionally, in this embodiment, the above electronic device can be at least one of multiple network devices in a computer network.

[0153] Optionally, in this embodiment, the above processor can be configured to execute the methods in the embodiments of the present application through the computer program.

[0154] Optionally, those of ordinary skill in the art can understand that Figure 8 the structure shown is only schematic, Figure 8 and it does not limit the structure of the above-mentioned electronic device. For example, the electronic device may further include more or fewer components (such as a network interface, etc.) than those shown Figure 8 in the figure, or have a different configuration from that shown Figure 8 in the figure.

[0155] Among them, the memory 802 can be used to store software programs and modules, such as the program instructions / modules corresponding to the management method and device of the system in the embodiments of the present application. The processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 802, that is, realizes the management method of the above-mentioned system. The memory 802 may include a high-speed random access memory, and may further include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memories. In some instances, the memory 802 may further include a memory remotely disposed relative to the processor 804, and these remote memories may be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, enterprise intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 802 may specifically but not limitedly be used to store the above-mentioned video data and other information. As an example, as Figure 8 shown, the above-mentioned memory 802 may but not limitedly include the acquisition module 602, the first generation module 604, and the second generation module 606 in the management device of the above-mentioned system. In addition, it may further include but not limited to other module units in the management device of the above-mentioned system, which will not be elaborated in this example.

[0156] Optionally, the above-mentioned transmission device 806 is used to receive or send data via a network. Specific examples of the above-mentioned network may include wired networks and wireless networks. In one instance, the transmission device 806 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers through a network cable so as to communicate with the Internet or a local area network. In one instance, the transmission device 806 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0157] In addition, the above-mentioned electronic device further includes: a display 808, which is used to display the above-mentioned video data; and a connection bus 810, which is used to connect each module component in the above-mentioned electronic device.

[0158] In other embodiments, the above terminal device or server may be a node in a distributed system. Among them, the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes in the form of network communication. Among them, the nodes can form a peer-to-peer network, and any form of computing device, such as electronic devices like servers and terminals, can become a node in the blockchain system by joining the peer-to-peer network.

[0159] According to one aspect of the present application, there is provided a computer-readable storage medium. The processor of the electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the system management methods provided in various optional implementation manners of the management aspect of the above system.

[0160] Optionally, in this embodiment, the above computer-readable storage medium may be set to store instructions for executing the methods in various embodiments of the present application.

[0161] Optionally, in this embodiment, those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware of the terminal device through a program. The program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0162] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0163] If the integrated unit in the above embodiments is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the above computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, 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 the storage medium and includes several instructions for causing one or more electronic devices to execute all or part of the steps of the methods described in various embodiments of the present application.

[0164] In the above embodiments of the present application, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0165] In several embodiments provided by this application, it should be understood that the disclosed application program can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, 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 displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0166] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0167] In addition, in each embodiment of this application, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0168] The above is only the preferred embodiment of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A method for processing video data, characterized in that Including: Obtain the shooting configuration information corresponding to the target shooting device. The shooting positions of the target shooting device include preset points and non-preset points. The shooting configuration information includes a first analysis rule corresponding to the preset points and a second analysis rule corresponding to the non-preset points. The first analysis rule includes scene features, and the second analysis rule does not include the scene features. In response to the target shooting device entering the preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point. In response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features.

2. The method according to claim 1, wherein The step of "In response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features" includes: In response to the target shooting device entering the non-preset point for shooting, use a preset recognition method to perform scene feature recognition on the acquisition area of the target shooting device to determine the second scene features. Determine the second analysis rule based on the second scene features, and perform target detection on the acquisition area of the target shooting device based on the second analysis rule to determine the second detection result.

3. The method according to claim 2, wherein The step of "In response to the target shooting device entering the non-preset point for shooting, use a preset recognition method to perform scene feature recognition on the acquisition area of the target shooting device to determine the second scene features" includes: In response to the target shooting device entering the non-preset point for shooting, use a first recognition method to recognize the acquisition area to determine the target demarcation line. Use a second recognition method to recognize the acquisition area to determine at least one area to be shot. Use the preset recognition method to perform scene feature recognition on the target shooting area to determine the second scene features, where the target shooting area represents the area determined based on the target demarcation line and the at least one area to be shot.

4. The method according to claim 1, wherein The step of "In response to the target shooting device entering the preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point" includes: Obtain the target identifier corresponding to the preset point entered by the target shooting device. Determine the first analysis rule and the first scene features based on the target identifier; generate a first detection result based on the first analysis rule and the first scene features.

5. The method according to claim 1, wherein The method further includes: Obtain the quality identifier corresponding to the first detection result or the second detection result. In the case where the quality identifier is the first quality identifier, determine the target analysis result based on the first detection result or the second detection result. When the quality identifier is the second quality identifier, determine the target analysis result based on the first detection result or the second detection result, and perform early warning reporting, where the quality level corresponding to the first quality identifier is higher than the quality level corresponding to the second quality identifier; When the quality identifier is the third quality identifier, discard the corresponding detection result and perform early warning reporting, where the quality level corresponding to the second quality identifier is higher than the quality level corresponding to the third quality identifier.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: In response to a change in the shooting position, perform scene recognition on the current shooting position corresponding to the target shooting device to obtain a target scene result, where the target scene result is used to determine the acquisition area of the target shooting device; Adaptive update the first analysis rule or the second analysis rule based on the target scene result.

7. The method according to claim 6, characterized in that, The adaptive update of the first analysis rule or the second analysis rule based on the target scene result includes: When the target scene result indicates a long-distance scene, set the first analysis rule or the second analysis rule to perform target detection on a first range, and set the detection threshold of the target detection to a first value, and update the first analysis rule or the second analysis rule; When the target scene result indicates a medium-distance scene, set the first analysis rule or the second analysis rule to perform target detection on a second range, and set the detection threshold of the target detection to a second value, and update the first analysis rule or the second analysis rule, where the second range is smaller than the first range and the second value is smaller than the first value; When the target scene result indicates a close-up scene, set the first analysis rule or the second analysis rule to perform target detection on a third range, and set the detection threshold of the target detection to a third value, and update the first analysis rule or the second analysis rule, where the third range is smaller than the second range and the third value is smaller than the second value.

8. A processing device for video data, characterized in that, It includes: An acquisition module, configured to acquire the shooting configuration information corresponding to the target shooting device, where the shooting position of the target shooting device includes a preset point and a non-preset point, the shooting configuration information includes a first analysis rule corresponding to the preset point and a second analysis rule corresponding to the non-preset point, the first analysis rule includes scene features, and the second analysis rule does not include the scene features; A first generation module, configured to, in response to the target shooting device entering the preset point for shooting, generate a first detection result based on the first analysis rule and the first scene features preset for the preset point; A second generation module, configured to, in response to the target shooting device entering the non-preset point for shooting, identify the second scene features captured by the target shooting device at the non-preset point, and generate a second detection result based on the second analysis rule and the second scene features.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein the computer program, when run by an electronic device, executes the method described in any one of claims 1 to 7.

10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

11. An electronic device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.

Images