Security and protection intelligence starting method and system, computer equipment and readable storage medium

Abstract

The invention relates to a security intelligence starting method and system, computer equipment and a readable storage medium. The method comprises the following steps: acquiring code stream data output by front-end acquisition equipment on each video channel in real time; for each video channel, analyzing the code stream data of the video channel, and determining a scene type corresponding to the video channel; and determining a to-be-opened target intelligence type of the video channel according to the scene type corresponding to the video channel, and opening the target intelligence type. By adopting the method, the intelligent degree of intelligent opening can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of video technology, in particular to an intelligent security opening method and system, a computer device and a readable storage medium. BACKGROUND

[0002] In the field of video monitoring, through an IP camera (IPC) accessing a network video recorder (NVR), intelligent opening can perform dynamic target detection, behavior recognition and abnormal event warning on monitoring video.

[0003] With the increase of NVR channels and intelligent channels, there are 32 channels / 64 channels / 128 channels, etc. For common security scenarios, multiple areas or places need to be monitored and managed. In this scenario, the number of NVRs required and the number of IPCs accessing the NVRs can be dozens to hundreds or even thousands. For this scenario, in the aspects of initial installation and debugging of devices, intelligent configuration after camera position changes, etc., the corresponding intelligence is opened in a manual configuration manner. Therefore, it is necessary to improve the intelligent degree of security intelligence opening. SUMMARY

[0004] Therefore, it is necessary to provide an intelligent security opening method and system capable of improving the intelligent degree and opening efficiency of intelligent opening, a computer device, a computer readable storage medium and a computer program product.

[0005] In a first aspect, the present application provides an intelligent security opening method, comprising:

[0006] Real-time acquisition of code stream data output by a front-end acquisition device on each video channel;

[0007] For each video channel, analyzing the code stream data of the video channel to determine the scene type corresponding to the video channel;

[0008] According to the scene type corresponding to each video channel, determining the target intelligent type to be opened for the corresponding video channel, and opening the target intelligent type.

[0009] In one embodiment, the analyzing the code stream data of the video channel to determine the scene type corresponding to the video channel comprises:

[0010] For each video channel, image processing is performed on the code stream data of the video channel to extract the scene features of the code stream data;

[0011] Identify a scene feature of the code stream data, and determine a scene type corresponding to the video channel.

[0012] In one of the embodiments, the determining of the target intelligent type to be started by the video channel according to the scene type corresponding to the video channel comprises:

[0013] According to the scene type corresponding to the video channel, if there is no intelligent type corresponding to the scene type, a preset scene type library is obtained; each preset scene type in the preset scene type library has a corresponding preset scene feature and a preset intelligent type;

[0014] The similarity between the scene feature and the preset scene feature of each preset scene type is determined.

[0015] According to the similarity, the target scene type corresponding to the video channel is determined from the preset scene type library, and the preset intelligent type corresponding to the target scene type is determined as the target intelligent type to be started by the video channel.

[0016] In one of the embodiments, the determining of the target intelligent type to be started by the video channel according to the scene type corresponding to the video channel comprises:

[0017] According to the scene type corresponding to the video channel, at least two candidate target intelligent types of the video channel are determined.

[0018] If there is an association between at least two candidate target intelligent types, it is determined that there are at least two target intelligent types to be started by the video channel.

[0019] In one of the embodiments, the starting of the target intelligent type comprises:

[0020] The first intelligent type supported by a front-end acquisition device on each video channel and the second intelligent type supported by a back-end server corresponding to each video channel are determined.

[0021] For each video channel, if the target intelligent type is included in the first intelligent type, the target intelligent type is started at the front-end acquisition device end.

[0022] If the target intelligent type is not included in the first intelligent type and is included in the second intelligent type, the target intelligent type is started at the back-end server end.

[0023] In one of the embodiments, after the target intelligent type is started at the back-end server end, the method further comprises:

[0024] Obtain the remaining computing resources of the backend server;

[0025] If the remaining computing resources are less than the preset resources, then the preset priority of each target intelligence type is obtained;

[0026] Based on the remaining computing resources and the preset priority of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of intelligence types to be shut down.

[0027] In one embodiment, after the target intelligence type is enabled on the backend server, the method further includes:

[0028] Obtain the environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server;

[0029] If the remaining computing resources are less than the preset resources, then the preset priority of each of the target intelligence types is updated according to the environmental data;

[0030] Based on the remaining computing resources and the updated priorities of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of intelligence types to be shut down.

[0031] In one embodiment, the method further includes:

[0032] Obtain the environmental data of the scene to which each bitstream data belongs;

[0033] Based on the environmental data, update the sensitivity of the intelligent functions enabled in the front-end acquisition device and / or the back-end server, and the intelligent functions enabled will operate according to the updated sensitivity.

[0034] Secondly, this application also provides a security smart activation method, the system including a front-end data acquisition device and a back-end server, the back-end server including a data acquisition module, a scene type recognition module, and a smart activation module, wherein:

[0035] The front-end acquisition device is used to acquire video data and encode the video data to obtain bitstream data;

[0036] The data acquisition module is used to acquire the bitstream data output by the front-end acquisition device on each video channel of the back-end server in real time.

[0037] The scene type identification module is used to analyze the bitstream data of each video channel to determine the scene type corresponding to the video channel.

[0038] The intelligent activation module is used to determine the target intelligent type to be activated for each video channel according to the scene type corresponding to each video channel, and activate the target intelligent type.

[0039] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0040] Real-time acquisition of bitstream data output by front-end acquisition devices on each video channel;

[0041] For each video channel, the bitstream data of the video channel is analyzed to determine the scene type corresponding to the video channel;

[0042] Based on the scene type corresponding to each video channel, determine the target intelligence type to be enabled for each corresponding video channel, and enable the target intelligence type.

[0043] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0044] Real-time acquisition of bitstream data output by front-end acquisition devices on each video channel;

[0045] For each video channel, the bitstream data of the video channel is analyzed to determine the scene type corresponding to the video channel;

[0046] Based on the scene type corresponding to each video channel, determine the target intelligence type to be enabled for each corresponding video channel, and enable the target intelligence type.

[0047] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0048] Real-time acquisition of bitstream data output by front-end acquisition devices on each video channel;

[0049] For each video channel, the bitstream data of the video channel is analyzed to determine the scene type corresponding to the video channel;

[0050] Based on the scene type corresponding to each video channel, determine the target intelligence type to be enabled for each corresponding video channel, and enable the target intelligence type.

[0051] The aforementioned security intelligence activation method, system, computer equipment, computer-readable storage medium, and computer program product acquire the bitstream data output by the front-end acquisition devices on each video channel in real time. They then identify the scene type corresponding to each video channel based on the acquired bitstream data, and subsequently determine the intelligence corresponding to that scene type. This real-time acquisition method allows for timely detection of changes in the location of the front-end acquisition devices, the addition of new front-end acquisition devices, and situations where intelligence is not activated. Furthermore, through image processing methods, they can determine the current scene based on the bitstream data acquired by the front-end acquisition devices, and thus determine the required intelligence. Compared to manual configuration, this method achieves rapid intelligence activation and improves activation efficiency, enabling adaptive activation of the corresponding intelligence. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is an application environment diagram of a method for enabling security intelligence in one embodiment;

[0054] Figure 2 This is a flowchart illustrating the method for enabling security intelligence in one embodiment;

[0055] Figure 3 This is a flowchart illustrating the intelligent adjustment method in one embodiment;

[0056] Figure 4 This is a flowchart illustrating the intelligent adjustment method in another embodiment;

[0057] Figure 5 This is a flowchart illustrating the method for activating security intelligence in another embodiment;

[0058] Figure 6 This is a structural block diagram of a security intelligence activation system in one embodiment;

[0059] Figure 7 This is a block diagram of the backend server in a security intelligence activation system in one embodiment.

[0060] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] Currently, backend NVR storage recorders generally support multiple front and rear intelligent functions. As the number of NVR channels and intelligent channels increases, it is necessary to manually enable and configure the corresponding intelligent functions according to the actual image environment, resulting in low intelligence and low efficiency in enabling intelligent functions.

[0063] To address this technical issue, a smart security activation method is proposed. This method automatically identifies the bitstream data output by each front-end acquisition device, determines the corresponding scene type, and enables rapid and intelligent activation of the field, while reducing the workload of manual configuration.

[0064] The method for enabling security intelligence provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, the front-end acquisition device 102 communicates with the back-end server 104 via a network. A data storage system can store the data that the back-end server 104 needs to process. The data storage system can be integrated onto the back-end server 104 or placed in the cloud or on another network server. The back-end server 104 acquires the bitstream data output by the front-end acquisition device 102 on each video channel; for each video channel, it analyzes the bitstream data to determine the scene type corresponding to the video channel; and based on the scene type corresponding to each video channel, it determines the target intelligent type to be activated for each corresponding video channel.

[0065] The front-end acquisition device 102 can be, but is not limited to, a device with image acquisition capabilities, such as different types of IPCs. The back-end server 104 can be different types of NVRs, independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing cloud computing services.

[0066] In one exemplary embodiment, such as Figure 2 As shown, a method for enabling smart security is provided, which can be applied to... Figure 1 This example uses the backend server as an example, with the front-end acquisition device being an IPC and the backend server being an NVR. The steps include 202 to 206. Wherein:

[0067] Step 202: Acquire the bitstream data output by the front-end acquisition device on each video channel in real time.

[0068] The bitstream data is obtained by encoding video data from the scene using front-end acquisition devices. It can be understood that the front-end acquisition device can be at least one of the following: a newly connected IPC, an IPC whose location has changed, or an IPC whose intelligence is not enabled. As an edge device, the IPC can support intelligence, which can be understood as a monitoring mode, enabling monitoring and data processing. Different types of intelligence support different types of intelligence, including, but not limited to, SMD (Surface Mount Technology), perimeter detection, license plate detection, face detection and comparison, floor calibration, people counting, queue management, and other front-end and back-end intelligence.

[0069] SMD (Smart Motion Detection): Consumes very few chip resources and can identify targets such as people, non-motorized vehicles, and motorized vehicles in the current scene; Perimeter: Detects intrusion into adjacent areas and regions, and triggers an intelligent alarm for intrusion into corresponding areas by people, vehicles, and non-motorized vehicles in the scene; Face Detection and Comparison: Suitable for scenarios with many faces, it detects facial attributes (e.g., masks, glasses, expressions, etc.) and coordinates in the scene, saves them, and compares them with a database to identify strangers; License Plate Detection: Suitable for scenarios with many vehicles, it detects vehicle attributes (model, color, logo), coordinates, and license plates in the scene and saves them; Floor Labeling: Marks the unit number and floor number of each building, suitable for high-altitude object throwing detection, etc.; People Counting: Used to detect the number of people entering and leaving a certain area, suitable for areas such as canteens and parks; Queue Management: Counts the time people spend queuing in a certain area, suitable for scenarios such as hospitals, canteens, and railway stations.

[0070] For example, to dynamically enable intelligence, the bitstream data output by the front-end acquisition devices on each video channel is acquired in real time. Real-time acquisition of the bitstream data output by the front-end acquisition devices on each video channel may include: real-time acquisition of IPC location; if a change in the location information of an IPC with intelligence enabled is detected, the scene corresponding to the IPC may change, and the corresponding intelligence may also need to be adjusted, i.e., acquiring the bitstream data output by the target IPC whose location has changed. If a newly added IPC is detected, it means that the corresponding intelligence of the IPC is not enabled, and the bitstream data output by the newly added IPC is acquired.

[0071] Step 204: For each video channel, analyze the bitstream data of the video channel to determine the scene type corresponding to the video channel.

[0072] The scene type can be determined based on scene features in the bitstream data. This determination can be achieved through a trained image recognition model or by comparing scene features with preset scene features for a predefined scene type. Scene feature extraction can be implemented using existing image processing algorithms, which will not be elaborated upon here. Each video channel is connected to its corresponding front-end acquisition device.

[0073] The scenarios include: scenarios requiring the identification of specific target objects, such as identifying people, non-motorized vehicles, and motorized vehicles at school intersections and playgrounds; scenarios requiring the monitoring of intrusion into areas, such as dormitory entrances, laboratory entrances, riverbanks, and the edges of campus walls; scenarios requiring the detection of facial attributes, such as identifying faces at dormitory entrances and laboratory entrances; scenarios requiring license plate detection, such as monitoring license plates at school gates; scenarios requiring floor marking, such as marking the apartment number and floor number of each building to prevent objects from being thrown from heights; scenarios requiring people counting, such as counting the number of people entering and leaving a certain area, such as in canteens and parks; and scenarios requiring the counting of queuing time within a certain area, such as counting queuing time at hospitals, canteens, and railway stations.

[0074] For example, for each video channel, the video channel's bitstream data is input into a trained image recognition model, which outputs the scene type corresponding to each bitstream data. It should be noted that the backend server obtains the bitstream data from the frontend acquisition device, decodes the bitstream data to obtain the corresponding image data, and analyzes the image data to determine the scene type.

[0075] Step 206: Based on the scene type corresponding to each video channel, determine the target intelligence type to be enabled for each corresponding video channel, and enable the target intelligence type.

[0076] Each scenario type has a corresponding intelligent type. For example, scenarios that need to identify a specified target object correspond to SMD, scenarios that need to monitor area intrusion and area intrusion correspond to perimeter, scenarios that need license plate detection correspond to license plate detection, scenarios that need floor calibration correspond to floor calibration, scenarios that need to count the queuing time of people in a certain area correspond to queue management, and scenarios that need to count the queuing time of people in a certain area correspond to people statistics.

[0077] Enabling video channel intelligence can be done either at the front-end acquisition device or at the back-end server. The specific decision depends on whether the intelligence types supported by the front-end acquisition device match the target intelligence type to be enabled, and / or the intelligence types supported by the back-end server and the remaining computing resources of the back-end server.

[0078] Optionally, if the intelligent types supported by the front-end acquisition device match the target intelligent type to be enabled, the target intelligent type to be enabled for each video channel is determined according to the scene type corresponding to each video channel. An enable command carrying the target intelligent type to be enabled is sent to the front-end acquisition device, which responds to the enable command and enables the target intelligent type. If the intelligent types supported by the front-end acquisition device do not match the target intelligent type to be enabled, the back-end server enables the target intelligent type. Furthermore, if the current back-end server's computing resources are insufficient, the back-end server sorts all target intelligent types to be enabled according to a preset priority and enables the intelligent type with the highest priority.

[0079] For example, if the analysis of the actual distributed bitstream data indicates that the current scene is a school gate or garage entrance with vehicles entering and exiting, then it is determined that intelligent license plate detection needs to be enabled in this environment. If the analysis of the actual distributed bitstream data indicates that the current scene is a cafeteria window, and the areas of the cafeteria window and seats are analyzed to determine the actual queuing area, then intelligent queue management needs to be enabled here, and rule lines for queue management will be automatically drawn. If the analysis of the actual distributed video indicates that the current scene is near a campus wall, then based on the analysis of the bitstream, the location of the campus wall is determined, a perimeter algorithm is enabled, and rule lines for perimeter intrusion are automatically drawn based on the algorithm's analysis results, thus preventing outsiders from climbing over the wall to intrude into the campus. If the analysis of the actual distributed video indicates that the current scene is a school dormitory or teaching building, and the algorithm detects and analyzes the location of the corresponding floor, then intelligent floor labeling is automatically enabled, and rule lines for floor labeling are drawn. If the analysis of the actual distributed video indicates that the current scene is a school dormitory entrance or laboratory entrance, then facial recognition is automatically enabled to prevent outsiders from entering.

[0080] In the aforementioned method for enabling intelligent video channels, the bitstream data output by the front-end acquisition devices on each video channel is acquired in real time. The acquired bitstream data is then used to identify the scene type corresponding to each video channel, and the corresponding intelligence is determined. The real-time acquisition method can promptly detect changes in the location of the front-end acquisition devices, the addition of new front-end acquisition devices, and situations where intelligence is not enabled. Furthermore, through image processing methods, the current scene can be determined based on the bitstream data acquired by the front-end acquisition devices, and the required intelligence can be determined. Compared to manual configuration, this method achieves rapid intelligent activation and improves activation efficiency, enabling adaptive activation of the corresponding intelligence.

[0081] In one exemplary embodiment, a method for intelligently determining video channels is provided, including the following methods:

[0082] Method 1: For each video channel, perform image processing on the video channel's bitstream data to extract scene features from the bitstream data; identify the scene features of the bitstream data to determine the scene type corresponding to the video channel; and determine the target intelligent type to be activated for the video channel based on the scene type corresponding to the video channel.

[0083] The method of extracting scene features from the bitstream data through image processing can be implemented using existing methods, which will not be elaborated here. This method can be understood as using image processing algorithms to directly identify the scene type corresponding to each bitstream data, and obtain the target intelligent type to be activated in the corresponding scene. On this basis, in order to further determine the accuracy of the scene type, the scene type determined in this step can be verified.

[0084] It is understandable that there are instances where image processing algorithms cannot identify the scene type corresponding to the bitstream data. This situation can be addressed using the following method two.

[0085] Method 2: Based on the scene type corresponding to the video channel, if there is no intelligent type corresponding to the scene type, obtain the preset scene type library; each preset scene type in the preset scene type library has corresponding preset scene features and preset intelligent types; determine the similarity between the scene features and the preset scene features of each preset scene type; determine the target scene type corresponding to the video channel from the preset scene type library based on the similarity, and determine the preset intelligent type corresponding to the target scene type as the target intelligent type to be activated for the video channel.

[0086] The preset scene type library stores scene images of different regions or locations. The preset scene type library can be determined by specifying the intelligent and rule areas to be enabled for the acquired scene image, and then performing image processing analysis and feature extraction algorithms on the scene image to obtain the preset scene features of each preset scene type.

[0087] For example, I-frame data is extracted from the bitstream data and independently decoded to reconstruct a clear static image. This static image is then analyzed and its features are extracted to obtain the corresponding scene features. The similarity between these scene features and the preset scene features of each preset scene type is calculated. If the similarity is greater than a preset similarity and the difference between the similarity and the preset similarity is within a preset range, then the scene type corresponding to the preset scene type with that similarity is taken as the actual scene type of the video channel. Based on the matching results, the intelligent type to be enabled and the rule area are sent to the backend service server or the frontend acquisition device. Furthermore, an experience matching tag is generated, which includes the enabled intelligent type, the similarity, and the corresponding channel.

[0088] In the above method, based on intelligent analysis to determine the scene type, a preset scene type library is introduced to avoid situations where the scene type cannot be determined. This method ensures the accuracy of the scene type of the video channel.

[0089] In one exemplary embodiment, the target intelligence type to be activated for the video channel is determined based on the scene type corresponding to the video channel, including:

[0090] Based on the scene type corresponding to the video channel, it is determined that the video channel has at least two candidate target intelligence types. If there is a correlation between the at least two candidate target intelligence types, then it is determined that the video channel has at least two target intelligence types to be activated. For example, for a high-altitude object throwing detection scenario, the video channel needs to activate two intelligence types: floor calibration and motion detection. Activating two related intelligence types in the same video channel can improve the channel's utilization rate. In this approach, two related intelligence types can be activated in the same video channel, improving resource utilization.

[0091] Based on the above determination of the target intelligence type to be enabled for each video channel, the following provides specific details on enabling intelligence. In an exemplary embodiment, it further includes:

[0092] Determine the first intelligent type supported by the front-end acquisition device on each video channel, and the second intelligent type supported by the back-end server corresponding to each video channel; for each video channel, if the first intelligent type includes the target intelligent type, then enable the target intelligent type on the front-end acquisition device; if the first intelligent type does not include the target intelligent type but the second intelligent type includes the target intelligent type, then enable the target intelligent type on the back-end server.

[0093] In cases where both the front-end data acquisition device and the back-end server have target intelligence types, the target intelligence type should be enabled on the front-end data acquisition device first. This approach can alleviate the resource pressure on the back-end server and ensure the stability of the device.

[0094] In the above method, the optimal smart activation end is determined based on the smart features supported by the backend server and the frontend acquisition device, as well as the target smart type to be activated for each video channel, thereby improving resource utilization.

[0095] Furthermore, considering the stability of the device after intelligent activation, the intelligence of each video channel needs to be adjusted according to the actual situation. In an exemplary embodiment, such as... Figure 3 As shown, an intelligent adjustment method is provided, including the following steps:

[0096] Step 302: Obtain the remaining computing resources of the backend server.

[0097] The remaining computing resources can be performance-related resources, such as at least one of CPU and memory utilization.

[0098] Step 304: If the remaining computing resources are less than the preset resources, then obtain the preset priority of each target intelligence type.

[0099] The preset priority of the target intelligence type can be set according to actual needs. If the remaining computing resources are less than the preset resources, it means that it is not possible to ensure the normal operation of all enabled target intelligence types, and only the normal operation of some intelligence types can be supported. In order to ensure the reliability of security to the greatest extent, it is necessary to disable some enabled target intelligence types based on the preset priority of the target intelligence types.

[0100] Step 306: Based on the remaining computing resources and the preset priority of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of intelligence types.

[0101] The types of intelligence to be disabled are determined based on priority; for example, they are selected from lowest to highest priority. The number of intelligence types to be disabled is determined based on remaining computing resources and the resources required to enable intelligence.

[0102] Optionally, after enabling intelligence, the performance of the NVR device is detected. If the CPU utilization rate is >90% or the memory utilization rate is >90%, it indicates that the remaining computing resources are less than the preset resources. Based on the remaining computing resources and the preset priority of each target intelligence type, the intelligence types to be turned off and the number of intelligence types to be turned off are determined, and the corresponding number of intelligence types to be turned off are turned off on the backend server.

[0103] Understandably, before activating the target intelligent type to be activated for each video channel, it is also possible to obtain the remaining computing resources of the backend server. If the remaining computing resources are less than the preset resources, the preset priority of each target intelligent type is obtained. Based on the remaining computing resources and the preset priority of each target intelligent type, the final target intelligent type to be activated is determined from the target intelligent types to be activated.

[0104] In the above method, based on the remaining computing resources of the backend server and the priority of each intelligence, the intelligence with lower priority is turned off in order to dynamically adjust the intelligence and ensure the stability of the device.

[0105] In another exemplary embodiment, such as Figure 4 As shown, an intelligent adjustment method is provided, including the following steps:

[0106] Step 402: Obtain the environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server.

[0107] Step 404: If the remaining computing resources are less than the preset resources, then update the preset priority of each target intelligence type according to the environmental data.

[0108] It's understandable that the priority and sensitivity requirements for intelligence vary under the same environment and in different time periods and weather conditions, necessitating dynamic adjustments. Environmental data includes time period data, weather data, brightness data, and ambient noise data. Optionally, if remaining computing resources are less than preset resources, priorities are assigned to different intelligent functions (such as face recognition, motion detection, and abnormal behavior analysis) based on current environmental characteristics (e.g., light intensity, number of moving objects, ambient noise), i.e., updating the preset priorities for each target intelligence type. For example, when the backend server is under heavy load, face detection can be prioritized during the day, motion detection at night, and infrared analysis can be automatically prioritized in night mode.

[0109] Furthermore, in an exemplary embodiment, environmental data of the scene to which each bitstream data belongs is acquired; based on the environmental data, the sensitivity of the enabled intelligent detection in the front-end acquisition device and / or back-end server is updated, and the enabled intelligent detection operates according to the updated sensitivity. For example, when adjusting the intelligent sensitivity within the same time period, during the morning rush hour, the sensitivity of intelligent detection such as vehicle detection and face detection entering and exiting the school is required to be higher, so the sensitivity of these intelligent detections can be increased to increase the detection accuracy; at night, when the requirements for intelligent detection such as vehicle detection and face detection are not high, the sensitivity can be dynamically reduced; furthermore, the saved performance and other resources are used for other higher priority intelligent detections.

[0110] This approach adjusts the sensitivity of intelligence based on environmental data, further improving resource utilization while ensuring detection accuracy.

[0111] Step 406: Based on the remaining computing resources and the updated priorities of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of types.

[0112] Among them, the intelligent types to be shut down can be low-priority intelligent types. For example, when the remaining computing resources are less than the preset resources, the priority of each target intelligent type is determined after updating the preset priority of each target intelligent type according to the environmental data. Based on the remaining computing resources and the updated priority of each target intelligent type, the intelligent types to be shut down and the number of intelligent types to be shut down are determined, and the intelligent types to be shut down corresponding to the number of intelligent types to be shut down are shut down on the backend server.

[0113] In the above method, after intelligent activation, the video channels with intelligent activation can be adjusted according to the environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server, thereby improving resource utilization.

[0114] In one exemplary embodiment, such as Figure 5 The diagram shown is a flowchart of a method for activating smart security systems, including the following steps:

[0115] Prioritize the first intelligent type supported by the front-end acquisition device and the second intelligent type supported by the back-end server to obtain all intelligent types after sorting. The back-end server connects to multiple front-end acquisition devices and obtains the bitstream data output by the front-end acquisition devices on each video channel in real time. For each video channel, the bitstream data of the video channel is analyzed to determine the scene type corresponding to the video channel. Based on the scene type corresponding to the video channel, the target intelligent type to be enabled for the video channel is determined.

[0116] If the target intelligent type to be activated for each video channel cannot be determined, a preset scene type library is obtained. Each preset scene type in the preset scene type library has corresponding preset scene features and preset intelligent types. The similarity between the scene features and the preset scene features of each preset scene type is determined. Based on the similarity, the target scene type corresponding to the video channel is determined from the preset scene type library, and the preset intelligent type corresponding to the target scene type is determined as the target intelligent type to be activated for the video channel. Furthermore, if the target intelligent type for a scene cannot be determined based on the preset scene type library, the corresponding target intelligent type is determined manually.

[0117] With each video channel having its corresponding target intelligent type enabled, the remaining computing resources of the backend server are obtained. If the remaining computing resources are less than the preset resources, the preset priority of each target intelligent type is obtained. Based on the remaining computing resources and the preset priority of each target intelligent type, the intelligent types to be disabled and the number of intelligent types to be disabled are determined, and the corresponding number of intelligent types to be disabled on the backend server are disabled. And / or,

[0118] Obtain the environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server; if the remaining computing resources are less than the preset resources, update the preset priority of each target intelligence type according to the environmental data; based on the remaining computing resources and the updated priority of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the corresponding number of intelligence types to be shut down on the backend server. And / or,

[0119] Obtain environmental data for the scene to which each bitstream data belongs; based on the environmental data, update the sensitivity of the intelligent functions already enabled in the front-end acquisition devices and / or back-end servers, and those already enabled will operate according to the updated sensitivity.

[0120] Furthermore, if a change is detected in the scene image data corresponding to the bitstream data, the currently enabled target intelligence type is disabled, and the step of acquiring the bitstream data output by the front-end acquisition devices on each video channel in real time is executed. If the scene image data corresponding to the bitstream data has not changed, the intelligence enabling process ends.

[0121] It should be noted that the specific implementation of this embodiment can be achieved in the manner defined above, and will not be elaborated here.

[0122] In the above embodiments, image recognition and comparison algorithms can quickly analyze the bitstream data acquired by the camera or match images in the library based on preset scene types to determine the intelligent type that needs to be enabled for the current scene, thus achieving rapid intelligent scene activation and reducing the workload of manual configuration. On this basis, when the front-end acquisition device is dynamically updated, the NVR device can perform intelligent analysis based on the addition of IPCs or the movement of IPCs to other areas, thereby dynamically updating and adjusting intelligent rules and reducing the workload of manual adjustment and reconfiguration. In addition, it can also make new adjustments to the priority of intelligence based on the external environment and different time periods, maximizing resource utilization and improving resource efficiency.

[0123] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0124] Based on the same inventive concept, this application also provides a security intelligence activation system for implementing the aforementioned security intelligence activation method. The solution provided by this system is similar to the implementation described in the above method; therefore, the specific limitations of the one or more security intelligence activation system embodiments provided below can be found in the limitations of the security intelligence activation method described above, and will not be repeated here.

[0125] In one exemplary embodiment, such as Figure 6 and Figure 7 As shown, a smart security activation system is provided. This system includes a front-end data acquisition device 602 and a back-end server 604. A communication connection is established between the front-end data acquisition device and the back-end server. The back-end server 604, as shown... Figure 7 As shown, it includes a data acquisition module 702, a scene type recognition module 704, and a smart activation module 706, wherein:

[0126] The front-end acquisition device 602 is used to acquire video data and encode the video data to obtain bitstream data.

[0127] The data acquisition module 702 is used to acquire the bitstream data output by the front-end acquisition devices on each video channel of the back-end server in real time.

[0128] The scene type recognition module 704 is used to analyze the bitstream data of each video channel to determine the scene type corresponding to the video channel.

[0129] The intelligent activation module 706 is used to determine the target intelligent type to be activated for each video channel based on the scene type corresponding to each video channel.

[0130] The aforementioned security intelligence activation system acquires real-time bitstream data from front-end acquisition devices on each video channel. It then identifies the scene type corresponding to each video channel based on the acquired bitstream data, and subsequently determines the intelligence corresponding to that scene type. This real-time acquisition method allows for timely detection of changes in the location of front-end acquisition devices, the addition of new front-end acquisition devices, and situations where intelligence is not activated. Furthermore, through image processing methods, it can determine the current scene based on the bitstream data acquired by the front-end acquisition devices, and thus determine the required intelligence. Compared to manual configuration, this method achieves rapid intelligence activation and improves activation efficiency, enabling adaptive activation of the corresponding intelligence.

[0131] In an exemplary embodiment, the scene type recognition module 704 is used to perform image processing on the bitstream data of each video channel to extract scene features from the bitstream data.

[0132] The scene characteristics of the bitstream data are identified to determine the scene type corresponding to the video channel.

[0133] In an exemplary embodiment, the scene type recognition module 704 is used to obtain a preset scene type library if there is no intelligent type corresponding to the scene type corresponding to the video channel; each preset scene type in the preset scene type library has a corresponding preset scene feature and preset intelligent type.

[0134] Determine the similarity between scene features and preset scene features for each preset scene type;

[0135] Based on similarity, the target scene type corresponding to the video channel is determined from the preset scene type library, and the preset intelligent type corresponding to the target scene type is determined as the target intelligent type to be activated for the video channel.

[0136] In an exemplary embodiment, the intelligent activation module 706 is used to determine, based on the scene type corresponding to the video channel, that there are at least two candidate target intelligent types for the video channel.

[0137] If there is a correlation between at least two candidate target intelligence types, then it is determined that there are at least two target intelligence types to be enabled in the video channel.

[0138] In an exemplary embodiment, the intelligent activation module 706 is used to determine the first intelligent type supported by the front-end acquisition device on each video channel, and the second intelligent type supported by the back-end server corresponding to each video channel.

[0139] For each video channel, if the first intelligent type includes the target intelligent type, then the target intelligent type is enabled on the front-end acquisition device.

[0140] If the first intelligence type does not include the target intelligence type but the second intelligence type does include the target intelligence type, then the target intelligence type is enabled on the backend server.

[0141] In an exemplary embodiment, the backend server 604 further includes an update module, which is used to obtain the remaining computing resources of the backend server.

[0142] If the remaining computing resources are less than the preset resources, then obtain the preset priority of each target intelligence type;

[0143] Based on the remaining computing resources and the preset priorities of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the corresponding number of intelligence types to be shut down on the backend server.

[0144] The update module is used to obtain environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server;

[0145] If the remaining computing resources are less than the preset resources, the preset priority of each target intelligence type will be updated based on the environmental data.

[0146] Based on the remaining computing resources and the updated priorities of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the corresponding number of intelligence types to be shut down on the backend server.

[0147] The update module is used to obtain environmental data of the scene to which each bitstream data belongs;

[0148] Based on environmental data, update the sensitivity of the intelligent functions enabled in the front-end data acquisition devices and / or back-end servers. The intelligent functions enabled will then operate according to the updated sensitivity.

[0149] The modules in the aforementioned intelligent security system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0150] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores bitstream data and intelligent related rules, among other data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a method for enabling intelligent security.

[0151] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0152] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0153] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0154] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments. It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data must comply with relevant regulations.

[0155] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0156] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0157] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for activating a smart security system, characterized in that, The method includes: Real-time acquisition of bitstream data output by front-end acquisition devices on each video channel; For each video channel, the bitstream data of the video channel is analyzed to determine the scene type corresponding to the video channel; Based on the scene type corresponding to the video channel, determine the target intelligence type to be enabled for the video channel, and enable the target intelligence type.

2. The method according to claim 1, characterized in that, For each video channel, the process of analyzing the bitstream data of that video channel to determine the scene type corresponding to that video channel includes: For each video channel, image processing is performed on the bitstream data of the video channel to extract scene features from the bitstream data; The scene characteristics of the bitstream data are identified to determine the scene type corresponding to the video channel.

3. The method according to claim 2, characterized in that, The step of determining the target intelligence type to be activated for the video channel based on the scene type corresponding to the video channel includes: Based on the scene type corresponding to the video channel, if there is no intelligent type corresponding to the scene type, a preset scene type library is obtained; each preset scene type in the preset scene type library has a corresponding preset scene feature and preset intelligent type; Determine the similarity between the scene features and the preset scene features of each preset scene type; Based on the similarity, the target scene type corresponding to the video channel is determined from the preset scene type library, and the preset intelligent type corresponding to the target scene type is determined as the target intelligent type to be activated for the video channel.

4. The method according to claim 1, characterized in that, The step of determining the target intelligence type to be activated for the video channel based on the scene type corresponding to the video channel includes: Based on the scene type corresponding to the video channel, it is determined that the video channel has at least two candidate target intelligence types; If there is a correlation between at least two of the candidate target intelligence types, then it is determined that there are at least two target intelligence types to be activated in the video channel.

5. The method according to any one of claims 1 to 4, characterized in that, Enabling the target intelligence type includes: Determine the first intelligent type supported by the front-end acquisition device on each video channel, and the second intelligent type supported by the back-end server corresponding to each video channel; For each video channel, if the first intelligent type includes the target intelligent type, then the target intelligent type is enabled on the front-end acquisition device. If the first intelligence type does not include the target intelligence type but the second intelligence type does include the target intelligence type, then the target intelligence type is enabled on the backend server.

6. The method according to claim 5, characterized in that, After enabling the target intelligence type on the backend server, the method further includes: Obtain the remaining computing resources of the backend server; If the remaining computing resources are less than the preset resources, then the preset priority of each target intelligence type is obtained; Based on the remaining computing resources and the preset priority of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of intelligence types to be shut down.

7. The method according to claim 5, characterized in that, After enabling the target intelligence type on the backend server, the method further includes: Obtain the environmental data of the scene to which each bitstream data belongs and the remaining computing resources of the backend server; If the remaining computing resources are less than the preset resources, then the preset priority of each of the target intelligence types is updated according to the environmental data; Based on the remaining computing resources and the updated priorities of each target intelligence type, determine the intelligence types to be shut down and the number of intelligence types to be shut down, and shut down the intelligence types to be shut down on the backend server corresponding to the number of intelligence types to be shut down.

8. The method according to claim 5, characterized in that, The method further includes: Obtain the environmental data of the scene to which each bitstream data belongs; Based on the environmental data, update the sensitivity of the intelligent functions enabled in the front-end acquisition device and / or the back-end server, and the intelligent functions enabled will operate according to the updated sensitivity.

9. A security intelligent unlocking system, characterized in that, The system includes a front-end data acquisition device and a back-end server. The back-end server includes a data acquisition module and a scene type recognition module, wherein: The front-end acquisition device is used to acquire video data and encode the video data to obtain bitstream data; The data acquisition module is used to acquire the bitstream data output by the front-end acquisition device on each video channel of the back-end server in real time. The scene type identification module is used to analyze the bitstream data of each video channel to determine the scene type corresponding to the video channel. The intelligent activation module is used to determine the target intelligent type to be activated for each video channel according to the scene type corresponding to each video channel, and activate the target intelligent type.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images