Video Surveillance Method, Device, Storage Medium and System Based on Scenario Linkage

Abstract

The present application provides a video surveillance method, device, storage medium and system based on scene linkage, which includes: determining the scene range according to the received alarm instruction and obtaining the scene parameter information contained in the alarm instruction; sending the scene parameter information to all robots within the scene range; receiving the target time difference fed back by each robot within the scene range; when there is a target time difference of not less than zero among the target time differences fed back by each robot within the scene range, determining the best target robot through the target time difference, so that the camera that initiates the alarm instruction and the target robot cooperate and control the target robot to execute a secondary confirmation instruction to verify the received alarm instruction. Finally, the verified alarm instruction is sent to the client. In this way, through the linkage between the camera and the robot, the frequency of false alarms and missed alarms can be effectively reduced, thereby improving the accuracy of the alarm information generated by the video surveillance.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence application technology, and in particular to a video surveillance method, device, storage medium and system based on scene linkage. Background Art

[0002] With the deployment and increasing intelligence of video surveillance infrastructure, scenario-based applications of AI (Artificial Intelligence) are gradually being promoted in digital life and digital governance, improving people's living standards and overall governance. The precision and recall rates of AI algorithms are technical indicators for measuring the level of AI application research and development, and are also quality indicators for determining whether AI products are easy to use, user-friendly, secure, and reliable in practical applications.

[0003] In existing technology solutions, AI cameras are typically deployed within video surveillance areas, and the monitoring platform deploys and configures corresponding AI algorithms based on the business scenario. When an event occurs in the scene, the AI camera, through AI algorithm reasoning, reports the event alarm to the platform when a certain threshold is reached. The platform then forwards the alarm message to the user. However, due to the lack of measures to address false alarms and missed alarms in the scene, the deployed AI algorithms have low precision and recall rates, which in turn reduces the accuracy of the alarm information generated by video surveillance. Summary of the Invention

[0004] The purpose of this application is to solve at least one of the above-mentioned technical defects, especially the technical defect in the existing technology that the accuracy and recall rate of the deployed AI algorithm are low due to the lack of processing measures for false alarms and missed alarms in the scene, thereby reducing the accuracy of the alarm information generated by video surveillance.

[0005] In a first aspect, the present application provides a scene-linked video surveillance method, which is applied to a monitoring platform in a scene-linked video surveillance system, the method comprising:

[0006] When receiving an alarm instruction sent by the camera, determine the scene range according to the alarm instruction and obtain the scene parameter information contained in the alarm instruction;

[0007] Sending the scene parameter information to all robots within the scene range to trigger each robot within the scene range to determine its corresponding target time difference according to the tailing effect time and the alarm occurrence position in the scene parameter information;

[0008] Receive target time difference feedback from each robot within the scene;

[0009] If there is a target time difference that is not less than zero among the target time differences fed back by each robot within the scene range, then, based on the target time differences corresponding to each robot within the scene range, the robot with the largest target time difference within the scene range is determined as the target robot;

[0010] issuing a secondary confirmation instruction to the target robot, and receiving response data corresponding to the secondary confirmation instruction fed back by the target robot;

[0011] The scene parameter information in the alarm instruction is verified according to the response data, and the verified alarm instruction is sent to the client.

[0012] In one embodiment, the checking of the scene parameter information in the alarm instruction includes:

[0013] Acquire a first scene type in the scene parameter information and a second scene type in the response data;

[0014] If the first scene type is inconsistent with the second scene type, modifying the first scene type in the scene parameter information according to the second scene type to obtain a new first scene type;

[0015] The alarm instruction is updated according to the new first scenario type to obtain a verified alarm instruction.

[0016] In one embodiment, the checking of the scene parameter information in the alarm instruction includes:

[0017] If the first scene type is consistent with the second scene type, the first scene type in the scene parameter information is not modified, and the alarm instruction is determined as a verified alarm instruction.

[0018] In one embodiment, the method further comprises:

[0019] If the target time difference feedbacked by each robot within the scene range is less than zero, feedback information indicating that secondary confirmation cannot be performed is generated and the feedback information is saved;

[0020] The alarm instruction is sent to the client.

[0021] In one embodiment, the method further comprises:

[0022] In each preset time period, the secondary confirmation efficiency corresponding to each scenario range in the preset time period is counted;

[0023] If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period is determined as the first target scene range, and the alarm threshold of each camera in each first target scene range is increased;

[0024] If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period will be determined as the second target scene range, and the alarm threshold of each camera in each second target scene range will be lowered.

[0025] In one embodiment, counting the secondary confirmation efficiency corresponding to each scenario range within the preset time period includes:

[0026] Obtaining a first target number corresponding to each scenario range; wherein the first target number is the number of feedback information generated within the preset time period and used to indicate that secondary confirmation cannot be made;

[0027] Counting the number of second targets corresponding to each scene range; wherein the number of second targets is the number of times the alarm instruction sent by the camera is received within the preset time period;

[0028] The difference between the number of second targets corresponding to each scene range and the number of first targets corresponding to it is used as the target difference corresponding to the scene range;

[0029] The quotient between the target difference corresponding to each scenario range and the second target quantity corresponding to each scenario range is determined as the secondary confirmation efficiency corresponding to the scenario range within the preset time period.

[0030] In one embodiment, the method further comprises:

[0031] Log data of a scene-linked video surveillance system is collected regularly, and the established intelligent algorithm model is updated based on the log data; wherein the intelligent algorithm model is used to manage the algorithm deployed by the monitoring terminal; the monitoring terminal includes a camera and a robot.

[0032] In a second aspect, the present application provides a scene-linked video surveillance device, which is applied to a monitoring platform in a scene-linked video surveillance system, and the device includes:

[0033] An alarm instruction receiving module is used to, when receiving an alarm instruction sent by a camera, determine the scene range according to the alarm instruction and obtain the scene parameter information contained in the alarm instruction;

[0034] a parameter information sending module, configured to send the scene parameter information to all robots within the scene range, so as to trigger each robot within the scene range to determine its corresponding target time difference according to the tailing effect time and the alarm occurrence position in the scene parameter information;

[0035] A target time difference receiving module is used to receive the target time difference feedback from each robot within the scene range;

[0036] a robot determination module configured to, if any target time difference among the target time differences fed back by each robot within the scene range is not less than zero, determine, based on the target time differences corresponding to each robot within the scene range, the robot with the largest target time difference within the scene range as the target robot;

[0037] a response data receiving module, configured to issue a secondary confirmation instruction to the target robot and receive response data corresponding to the secondary confirmation instruction fed back by the target robot;

[0038] The alarm instruction sending module is used to verify the scene parameter information in the alarm instruction according to the response data, and send the verified alarm instruction to the client.

[0039] In a third aspect, the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the scene-linked video surveillance method as described in any of the above embodiments.

[0040] In a fourth aspect, the present application provides a video surveillance system based on scene linkage, the system comprising a monitoring platform and a monitoring terminal, the monitoring terminal comprising at least one camera and a robot;

[0041] The monitoring platform is used to execute the steps of the scene-linked video monitoring method as described in any of the above embodiments.

[0042] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0043] The present application provides a video surveillance method, device, storage medium and system based on scene linkage, which includes: when receiving an alarm command sent by a camera, determining the scene range according to the alarm command and obtaining the scene parameter information contained in the alarm command; sending the scene parameter information to all robots within the scene range; receiving the target time difference feedback from each robot within the scene range; when there is a target time difference not less than zero among the target time differences feedback from each robot within the scene range, determining the best target robot based on the target time difference, so that the camera that initiates the alarm command and the target robot cooperate, and control the target robot to execute a secondary confirmation command to verify the received alarm command. Finally, the verified alarm command is sent to the client. In this way, through the linkage between the camera and the robot, the frequency of false alarms and missed alarms can be effectively reduced, thereby improving the accuracy of the alarm information generated by the video surveillance. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 A flowchart of a scene-based video surveillance method provided in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of a process for verifying scene parameter information in an alarm instruction provided in an embodiment of the present application;

[0047] Figure 3 A timing diagram of a scene-linked video surveillance method provided in an embodiment of the present application;

[0048] Figure 4 A schematic diagram of the structure of a video surveillance device based on scene linkage provided in an embodiment of the present application;

[0049] Figure 5 A schematic structural diagram of a scene-linked video surveillance system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0051] In one embodiment, the present application provides a video surveillance method based on scene linkage. The following embodiment is described by applying this method to a monitoring platform in a video surveillance system based on scene linkage.

[0052] like Figure 1 As shown, the present application provides a scene-linked video surveillance method, which is applied to a monitoring platform in a scene-linked video surveillance system, and the method includes:

[0053] Step S101: When an alarm instruction is received from a camera, the scene range is determined according to the alarm instruction, and scene parameter information included in the alarm instruction is obtained.

[0054] The scene parameter information includes the scene type, the tailing effect time and the alarm occurrence location.

[0055] In this step, when the camera detects a scenario in its deployed algorithm, it generates an alarm based on the scenario type, the corresponding tailing effect duration, and the location of the scenario (the alarm location). The alarm is then sent to the monitoring platform. Upon receiving the alarm, the monitoring platform determines the scope of the scenario based on the scenario parameters in the alarm, and retrieves the scenario parameters contained in the alarm.

[0056] It is understood that the tailing effect time refers to the time interval from the triggering of the alarm in the corresponding scene to the restoration of the scene to normal. The tailing effect time for different scenes can be pre-set or determined according to preset rules. This application does not impose specific restrictions on this.

[0057] Furthermore, the scene range is determined based on the scene parameter information in the alarm instruction, which may specifically include: searching for the scene range corresponding to the alarm location in the preset area division list based on the alarm location in the scene parameter information to determine the scene range corresponding to the alarm instruction.

[0058] Step S102: Send the scene parameter information to all robots within the scene range to trigger each robot within the scene range to determine its corresponding target time difference according to the tailing effect time and alarm occurrence position in the scene parameter information.

[0059] In this step, upon receiving an alarm command, acquiring scene parameter information, and determining the scene range, the monitoring platform sends the scene parameter information to all robots within the determined scene range. When a robot within the scene range receives the information sent by the monitoring platform, it determines its distance to the alarm location based on the alarm location in the scene parameter information. Combined with its average movement speed, it calculates the time required for each robot within the scene range to move to the alarm location. The robot then subtracts the tailing effect time in the scene parameter information from the sum of the time required for each robot within the scene range to move to the alarm location and the time required for the robot to perform a secondary confirmation. This allows each robot within the scene range to determine its corresponding target time difference.

[0060] It can be understood that the purpose of triggering each robot within the scene range to determine its corresponding target time difference based on the tail effect time and the alarm occurrence location in the scene parameter information is to enable the monitoring platform to determine whether there is a robot among the robots within the scene range that can reach the alarm occurrence location within the effective time for secondary confirmation.

[0061] Step S103: Receive the target time difference fed back by each robot within the scene range.

[0062] In this step, after the monitoring platform triggers each robot within the scene to determine its corresponding target time difference based on the trailing effect time and alarm occurrence location in the scene parameter information, it needs to wait for each robot within the scene to feedback the target time difference. This allows the monitoring platform to determine whether secondary confirmation is required and, if so, which target robot to assign for secondary confirmation.

[0063] Step S104: Determine whether the target time difference fed back by each robot within the scene range is less than zero.

[0064] It is understandable that when the target time difference feedback from each robot within the scene range is received, it is necessary to judge whether the target time difference feedback from each robot within the scene range is less than zero. For the robot, if its corresponding target time difference is less than zero, it means that the robot cannot reach the alarm location within the effective time for secondary confirmation.

[0065] Step S105: If there is a target time difference that is not less than zero among the target time differences fed back by each robot within the scene range, the robot with the largest target time difference within the scene range is determined as the target robot based on the target time differences corresponding to each robot within the scene range.

[0066] In this step, if the target time difference reported by each robot within the scene range is not less than zero, it indicates that one of the robots within the scene range can reach the alarm location for secondary confirmation within the effective time. At this point, the robot with the largest target time difference within the scene range is determined as the target robot. This allows the robot to reach the alarm location for secondary confirmation as quickly as possible, while also allowing the monitoring platform to receive feedback from the target robot more quickly, thereby shortening the time interval between triggering and receiving an alarm in the video surveillance method. This improves the monitoring efficiency of the scene-linked video surveillance method.

[0067] For example, assuming that the tailing effect time corresponding to scene type A is 30s, there are three robots within the determined scene range, represented by robot a, robot b and robot c. The time required for robots a, robot b and robot c to reach the alarm location is 12s, 16s and 18s respectively, and the time required for secondary confirmation is 12s, 15s and 5s respectively. At this time, robots a, robot b and robot c respectively calculate their own target time differences, that is, the target time difference of robot a is 6s (30-(12+12)), the target time difference of robot b is -1s (30-(15+16)), and the target time difference of robot c is 7s (30-(18+5)). At this time, the target time difference of the robots within the scene range is not less than zero. Therefore, it is necessary to issue a secondary confirmation instruction, determine the robot with the largest target time difference (robot c) as the target robot, and issue a secondary confirmation instruction to the target robot.

[0068] Step S106: Send a secondary confirmation instruction to the target robot, and receive response data corresponding to the secondary confirmation instruction fed back by the target robot.

[0069] The secondary determination instruction includes the location where the alarm occurs.

[0070] Once the monitoring platform identifies the target robot, it sends a secondary confirmation command to the target robot, directing it to the location where the alarm occurred for secondary confirmation. The platform then waits for the target robot to respond with the corresponding secondary confirmation data, allowing the monitoring platform to determine whether the alarm command initiated by the camera needs to be corrected.

[0071] Step S107: According to the response data, the scene parameter information in the alarm instruction is verified, and the verified alarm instruction is sent to the client.

[0072] In this step, when the monitoring platform receives the response data, it compares the response data with the scene parameter information in the received alarm instruction to determine whether the alarm instruction needs to be revised. When the verification of the scene parameter information in the alarm instruction is completed, the verified alarm instruction can be sent to the client. It is understandable that the verified alarm instruction may be the same as the original alarm instruction, or it may be different, depending on whether the monitoring platform revised the alarm instruction during the process of verifying the scene parameter information in the alarm instruction.

[0073] Step S108: If the target time differences fed back by each robot within the scene range are all less than zero, feedback information indicating that secondary confirmation cannot be performed is generated and the feedback information is saved.

[0074] If the target time difference reported by every robot within the scene is less than zero, it means that none of the robots within the scene can reach the alarm location for secondary confirmation within the valid time. In this case, the monitoring platform does not issue a secondary confirmation instruction to any robot within the scene. Instead, it directly generates feedback indicating that secondary confirmation is impossible. This prevents the robots within the scene from wasting effort and wasting resources.

[0075] Step S109: Send the alarm instruction to the client.

[0076] It is understandable that when all robots within the scene range are unable to reach the alarm location for secondary confirmation within the effective time, there is no secondary confirmation result, and the alarm instruction sent by the camera can be directly sent to the client.

[0077] The present application provides a video surveillance method, device, storage medium and system based on scene linkage, which includes: when receiving an alarm command sent by a camera, determining the scene range according to the alarm command and obtaining the scene parameter information contained in the alarm command; sending the scene parameter information to all robots within the scene range; receiving the target time difference feedback from each robot within the scene range; when there is a target time difference not less than zero among the target time differences feedback from each robot within the scene range, determining the best target robot based on the target time difference, so that the camera that initiates the alarm command and the target robot cooperate, and control the target robot to execute a secondary confirmation command to verify the received alarm command. Finally, the verified alarm command is sent to the client. In this way, through the linkage between the camera and the robot, the frequency of false alarms and missed alarms can be effectively reduced, thereby improving the accuracy of the alarm information generated by the video surveillance.

[0078] like Figure 2As shown, in one embodiment, checking the scene parameter information in the alarm instruction includes:

[0079] Step S201: Acquire the first scene type in the scene parameter information and the second scene type in the response data.

[0080] The first scene type refers to the scene type in the scene parameter information, and the second scene type refers to the scene type in the response data fed back during the second confirmation by the target robot.

[0081] In this step, when checking the scene parameter information in the alarm instruction, it is necessary to obtain the object to be checked, that is, the first scene type in the scene parameter information and the second scene type in the response data.

[0082] Step S202: Determine whether the first scene type is consistent with the second scene type.

[0083] The acquired first scene type is compared with the second scene type, and it is determined whether the first scene type is consistent with the second scene type.

[0084] Step S203: If the first scene type is inconsistent with the second scene type, the first scene type in the scene parameter information is modified according to the second scene type to obtain a new first scene type;

[0085] When the first scene type and the second scene type are inconsistent, the monitoring platform will modify the first scene type in the scene parameter information based on the second scene type. For example, if the first scene type is a smoking scene and the second scene type is a fire scene, the first scene type will be modified to a fire scene to obtain a new first scene type. If the first scene type is a smoking scene and the second scene type is also a smoking scene, there is no need to modify the first scene type.

[0086] Step S204: updating the alarm instruction according to the new first scenario type to obtain a verified alarm instruction.

[0087] In this step, when a new first scene type is obtained, the scene type in the scene parameter information in the alarm instruction can be updated to the new first scene type to complete the update of the alarm instruction and obtain a verified alarm instruction.

[0088] Step S205: If the first scene type is consistent with the second scene type, the first scene type in the scene parameter information is not modified, and the alarm instruction is determined as a verified alarm instruction.

[0089] It is understood that the received alarm instructions are verified for secondary confirmation. Ultimately, the verified alarm instructions are sent to the client. This linkage between the camera and the robot allows for secondary confirmation of the alarm instructions, effectively reducing the frequency of false alarms and missed alarms, thereby improving the accuracy of alarm information generated by video surveillance.

[0090] In one embodiment, the scene-linked video surveillance method further includes:

[0091] In each preset time period, the secondary confirmation efficiency corresponding to each scenario range in the preset time period is counted;

[0092] If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period is determined as the first target scene range, and the alarm threshold of each camera in each first target scene range is increased;

[0093] If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period will be determined as the second target scene range, and the alarm threshold of each camera in each second target scene range will be lowered.

[0094] In this embodiment, the preset time period can be determined by a preset period determination rule or set by staff. Within each preset time period, the monitoring platform will collect statistics on the secondary confirmation efficiency of each corresponding scene range. When the secondary confirmation efficiency corresponding to each scene range in the current preset time period is compared with the secondary confirmation efficiency of the scene range in the previous preset time period, if the secondary confirmation efficiency of the scene range shows an upward or downward trend, the alarm threshold of each camera within the scene range will be adjusted. The direction of adjustment depends on whether the secondary confirmation efficiency of the scene range shows an upward or downward trend.

[0095] Furthermore, adjusting the alarm threshold of each camera within the scene range can specifically include: determining the adjustment step of each camera within the scene range according to the preset adjustment rules and the current alarm threshold of each camera within the scene range; and raising or lowering the alarm threshold of each camera within the scene range according to the adjustment step of each camera within the scene range.

[0096] It can be understood that within each preset time period, statistics on the secondary confirmation efficiency corresponding to each scene range within the preset time period can realize dynamic adjustment of the alarm threshold of the camera within each scene range, thereby adjusting the frequency of alarm triggering, improving the recall rate of the deployed artificial intelligence algorithm, and then saving the resources of the monitoring terminal within a reasonable range to improve the resource utilization of the terminal equipment.

[0097] In one embodiment, counting the secondary confirmation efficiency corresponding to each scenario range within the preset time period includes:

[0098] Obtaining a first target number corresponding to each scenario range; wherein the first target number is the number of feedback information generated within the preset time period and used to indicate that secondary confirmation is impossible;

[0099] Counting the number of second targets corresponding to each scene range; wherein the number of second targets is the number of times the alarm instruction sent by the camera is received within the preset time period;

[0100] The difference between the number of second targets corresponding to each scene range and the number of first targets corresponding to it is used as the target difference corresponding to the scene range;

[0101] The quotient between the target difference corresponding to each scenario range and the second target quantity corresponding to each scenario range is determined as the secondary confirmation efficiency corresponding to the scenario range within the preset time period.

[0102] In this embodiment, the secondary confirmation efficiency corresponding to each scenario range within the preset time period can be calculated according to the following expression:

[0103]

[0104] Where, The secondary confirmation efficiency corresponding to each scenario range is is the second target number corresponding to each scene range, that is, the number of times each scene range receives the alarm instruction sent by the camera within the preset time period, It is the first target number corresponding to each scenario range, that is, the number of feedback information generated by each scenario range within the preset time period and used to indicate that secondary confirmation cannot be made.

[0105] In one embodiment, the scene-linked video surveillance method further includes:

[0106] Log data of the scene-based video surveillance system is collected regularly, and the established intelligent algorithm model is updated based on the log data; the intelligent algorithm model is used to manage the algorithms deployed on the monitoring terminals; the monitoring terminals include cameras and robots.

[0107] In this embodiment, log data refers to data recording events, state changes, user activities, and other information within a scene-based video surveillance system. It can be used to track system operation and usage. It is understood that by updating the intelligent algorithm model, different algorithms can be deployed for different scene ranges, making it more targeted. This can effectively reduce the frequency of false positives and missed negatives and improve the accuracy of alarm information generated by video surveillance.

[0108] In one example, if Figure 3 As shown, this application provides a timing diagram of a video surveillance method based on scene linkage. Figure 3 The system includes four main parts: a robot, a camera, a monitoring platform, and a client. On the one hand, when the scene algorithm deployed in the camera is running, the camera detects that a scene event has occurred, and the scene parameter information is sent to the monitoring platform as part of the alarm instruction. When the monitoring platform receives the alarm instruction, the alarm linkage is triggered, that is, the robot within the corresponding scene range is controlled to calculate its corresponding target time difference, and the target time difference fed back by the robot is received, and then the target robot is determined, and a secondary confirmation instruction is issued to the target robot so that the target robot goes to the location where the alarm occurred for on-site secondary confirmation. When the secondary confirmation result fed back by the target robot is received, the alarm instruction is verified according to the secondary confirmation result, and the verified alarm instruction is pushed to the client. On the other hand, when the monitoring platform detects that the preset conditions are met, it adjusts the alarm threshold of the camera. It is understandable that this method can be applied to various intelligent applications and services for video surveillance, such as Tianyi Home Security Application, Tianyi Cloud Eye products, etc.

[0109] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0110] The following describes a video surveillance device based on scene linkage provided in an embodiment of the present application. The video surveillance device based on scene linkage described below and the video surveillance method based on scene linkage described above can refer to each other.

[0111] like Figure 4 As shown, the present application provides a scene-linked video surveillance device 300, which is applied to a monitoring platform in a scene-linked video surveillance system. The device includes:

[0112] The alarm instruction receiving module 301 is used to determine the scene range according to the alarm instruction when receiving the alarm instruction sent by the camera, and obtain the scene parameter information contained in the alarm instruction;

[0113] A parameter information sending module 302 is configured to send the scene parameter information to all robots within the scene range, so as to trigger each robot within the scene range to determine its corresponding target time difference according to the tailing effect time and the alarm occurrence position in the scene parameter information;

[0114] The target time difference receiving module 303 is used to receive the target time difference feedback from each robot within the scene range;

[0115] The robot determination module 304 is configured to determine, if any of the target time differences reported by each robot within the scene range is not less than zero, the robot with the largest target time difference within the scene range as the target robot based on the target time differences corresponding to each robot within the scene range;

[0116] The response data receiving module 305 is used to send a secondary confirmation instruction to the target robot and receive response data corresponding to the secondary confirmation instruction fed back by the target robot;

[0117] The alarm instruction sending module 306 is used to verify the scene parameter information in the alarm instruction according to the response data, and send the verified alarm instruction to the client.

[0118] In one embodiment, the alarm instruction sending module includes:

[0119] A scene type acquisition submodule, configured to acquire the first scene type in the scene parameter information and the second scene type in the response data;

[0120] a first judgment submodule, configured to, if the first scene type is inconsistent with the second scene type, modify the first scene type in the scene parameter information according to the second scene type to obtain a new first scene type;

[0121] The alarm instruction updating submodule is used to update the alarm instruction according to the new first scenario type to obtain a verified alarm instruction.

[0122] In one embodiment, the alarm instruction sending module includes:

[0123] The second judgment submodule is configured to: if the first scene type is consistent with the second scene type, not to modify the first scene type in the scene parameter information, and to determine the alarm instruction as a verified alarm instruction.

[0124] In one embodiment, the scene-linked video surveillance device further includes:

[0125] A feedback information generation module is used to generate feedback information indicating that secondary confirmation cannot be performed if the target time difference feedback from each robot within the scene range is less than zero, and save the feedback information;

[0126] The instruction sending module is used to send the alarm instruction to the client.

[0127] In one embodiment, the scene-linked video surveillance device further includes:

[0128] A statistics module is used to count the secondary confirmation efficiency of each scenario range within each preset time period;

[0129] A first threshold adjustment module is configured to, if there is a scene range whose corresponding secondary confirmation efficiency rate shows a continuous downward trend within the preset time period, determine the scene range whose corresponding secondary confirmation efficiency rate shows a continuous downward trend within the preset time period as a first target scene range, and increase the alarm threshold of each camera in each first target scene range;

[0130] The second threshold adjustment module is used to determine the scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period as the second target scene range if there is a scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period, and lower the alarm threshold of each camera in each second target scene range.

[0131] In one embodiment, the statistics module includes:

[0132] A data acquisition submodule is configured to acquire a first target number corresponding to each scenario range; wherein the first target number is the number of feedback information generated within the preset time period and used to indicate that secondary confirmation is impossible;

[0133] A data statistics submodule is used to count the number of second targets corresponding to each scene range; wherein the second target number is the number of times the alarm instruction sent by the camera is received within the preset time period;

[0134] a target difference determination submodule, configured to take the difference between the second target quantity corresponding to each scene range and the first target quantity corresponding thereto as the target difference corresponding to the scene range;

[0135] The efficiency determination submodule is used to determine the quotient between the target difference corresponding to each scene range and its corresponding second target quantity as the secondary confirmation efficiency corresponding to the scene range within the preset time period.

[0136] In one embodiment, the scene-linked video surveillance device further includes:

[0137] The model update module is used to regularly collect log data from the scene-linked video surveillance system and update the established intelligent algorithm model based on the log data; the intelligent algorithm model is used to manage the algorithms deployed by the monitoring terminals; the monitoring terminals include cameras and robots.

[0138] The division of the various modules in the above-mentioned scene-based linkage video surveillance device is for illustrative purposes only. In other embodiments, the scene-based linkage video surveillance device can be divided into different modules as needed to complete all or part of the functions of the above-mentioned scene-based linkage video surveillance device. The various modules in the above-mentioned scene-based linkage video surveillance device can be implemented in whole or in part through software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of the above-mentioned modules.

[0139] In one embodiment, the present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the scene-linked video surveillance method as described in any of the above embodiments.

[0140] In one embodiment, the present application also provides a video surveillance system based on scene linkage, the system including a monitoring platform and a monitoring terminal, the monitoring terminal including at least one camera and a robot; the monitoring platform is used to execute the steps of the video surveillance method based on scene linkage as described in any of the above embodiments.

[0141] Schematically, as Figure 5 As shown, Figure 5 A video surveillance system based on scene linkage is provided in an embodiment of the present application.

[0142] (1) Client: refers to various terminal functional software used by video surveillance users, including but not limited to APP, mini-programs, H5 Web, etc., which has functions such as viewing surveillance videos (live broadcast, playback), receiving alarm commands, etc.

[0143] (2) Monitoring platform: It can be applied to computer equipment. The monitoring platform includes an algorithm management module, an intelligent algorithm model, and an alarm message center, among which:

[0144] Algorithm management module: Use intelligent algorithm models to manage the artificial intelligence algorithms of monitoring scenarios, deploy algorithms on cameras, and configure various parameters of algorithm reasoning, including setting and adjusting alarm thresholds.

[0145] Intelligent algorithm model: The artificial intelligence algorithm is trained based on big data to form an artificial intelligence algorithm model, and the system log data is continuously collected to update the big data and then correct the algorithm model.

[0146] Alarm Message Center: Receives alarm commands reported by cameras and secondary confirmation results reported by robots, and pushes linkage messages to robots to trigger them to calculate target time difference.

[0147] (3) Monitoring terminals: including cameras and robots. It can be understood that cameras and robots are both intelligent devices with data calculation and processing capabilities.

[0148] Camera: Accepts the artificial intelligence algorithm deployment and parameter configuration of the monitoring platform, runs the artificial intelligence algorithm reasoning operation, and triggers the alarm instruction report when the operation result reaches its corresponding alarm threshold.

[0149] Robot: When receiving the linkage message (scene parameter information) sent by the monitoring platform, it calculates the corresponding target time difference. When receiving the secondary confirmation instruction, it quickly moves to the corresponding alarm location; and runs the corresponding algorithm to conduct on-site secondary confirmation and report the secondary confirmation results.

[0150] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the scene-linked video surveillance system to which the solution of the present application is applied. The specific scene-linked video surveillance system may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0151] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements not only include those elements, but also include other elements not clearly listed, or also include elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or equipment comprising the elements. Herein, the singular forms "one", "an" and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" etc. specify the existence of stated features, wholes, steps, operations, components, parts or combinations thereof, but do not exclude the possibility of the existence or addition of one or more other features, wholes, steps, operations, components, parts or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0152] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0153] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A video surveillance method based on scene linkage, characterized in that: The method is applied to a monitoring platform in a scene-linked video monitoring system, and includes: When receiving an alarm instruction sent by the camera, determine the scene range according to the alarm instruction and obtain the scene parameter information contained in the alarm instruction; Sending the scene parameter information to all robots within the scene range to trigger each robot within the scene range to determine its corresponding target time difference based on the tail effect time and alarm occurrence location in the scene parameter information; wherein the tail effect time refers to the time interval from the triggering of the alarm in its corresponding scene to the restoration of the scene to normal; Receive target time difference feedback from each robot within the scene; If there is a target time difference that is not less than zero among the target time differences fed back by each robot within the scene range, then, based on the target time differences corresponding to each robot within the scene range, the robot with the largest target time difference within the scene range is determined as the target robot; issuing a secondary confirmation instruction to the target robot, and receiving response data corresponding to the secondary confirmation instruction fed back by the target robot; Verifying the scene parameter information in the alarm instruction according to the response data, and sending the verified alarm instruction to the client; In each preset time period, the secondary confirmation efficiency corresponding to each scenario range in the preset time period is counted; If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous downward trend within the preset time period is determined as the first target scene range, and the alarm threshold of each camera in each first target scene range is increased; If there is a scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period, the scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period will be determined as the second target scene range, and the alarm threshold of each camera in each second target scene range will be lowered.

2. The video surveillance method based on scene linkage according to claim 1 is characterized in that: The checking of the scene parameter information in the alarm instruction includes: Acquire a first scene type in the scene parameter information and a second scene type in the response data; If the first scene type is inconsistent with the second scene type, modifying the first scene type in the scene parameter information according to the second scene type to obtain a new first scene type; The alarm instruction is updated according to the new first scenario type to obtain a verified alarm instruction.

3. The video surveillance method based on scene linkage according to claim 2 is characterized in that: The checking of the scene parameter information in the alarm instruction includes: If the first scene type is consistent with the second scene type, the first scene type in the scene parameter information is not modified, and the alarm instruction is determined as a verified alarm instruction.

4. The video surveillance method based on scene linkage according to claim 1 is characterized in that: The method further comprises: If the target time difference feedbacked by each robot within the scene range is less than zero, feedback information indicating that secondary confirmation cannot be performed is generated and the feedback information is saved; The alarm instruction is sent to the client.

5. The video surveillance method based on scene linkage according to claim 1 is characterized in that: Statistics on the secondary confirmation efficiency of each scenario range within the preset time period, including: Obtaining a first target number corresponding to each scenario range; wherein the first target number is the number of feedback information generated within the preset time period and used to indicate that secondary confirmation cannot be made; Counting the number of second targets corresponding to each scene range; wherein the number of second targets is the number of times the alarm instruction sent by the camera is received within the preset time period; The difference between the number of second targets corresponding to each scene range and the number of first targets corresponding to it is used as the target difference corresponding to the scene range; The quotient between the target difference corresponding to each scenario range and the second target quantity corresponding to each scenario range is determined as the secondary confirmation efficiency corresponding to the scenario range within the preset time period.

6. The video surveillance method based on scene linkage according to claim 1 is characterized in that: The method further comprises: Log data of a scene-linked video surveillance system is collected regularly, and the established intelligent algorithm model is updated based on the log data; wherein the intelligent algorithm model is used to manage the algorithm deployed by the monitoring terminal; the monitoring terminal includes a camera and a robot.

7. A video surveillance device based on scene linkage, characterized in that: A monitoring platform used in a scene-linked video monitoring system, the device comprising: An alarm instruction receiving module is used to, when receiving an alarm instruction sent by a camera, determine the scene range according to the alarm instruction and obtain the scene parameter information contained in the alarm instruction; a parameter information sending module, configured to send the scene parameter information to all robots within the scene range, thereby triggering each robot within the scene range to determine its corresponding target time difference based on the tailing effect time and alarm occurrence location in the scene parameter information; wherein the tailing effect time refers to the time interval from the triggering of the alarm for the corresponding scene to the restoration of the scene to normal; A target time difference receiving module is used to receive the target time difference feedback from each robot within the scene range; a robot determination module configured to, if any target time difference among the target time differences fed back by each robot within the scene range is not less than zero, determine, based on the target time differences corresponding to each robot within the scene range, the robot with the largest target time difference within the scene range as the target robot; a response data receiving module, configured to issue a secondary confirmation instruction to the target robot and receive response data corresponding to the secondary confirmation instruction fed back by the target robot; an alarm instruction sending module, configured to verify the scene parameter information in the alarm instruction according to the response data, and send the verified alarm instruction to the client; A statistics module is used to count the secondary confirmation efficiency of each scenario range within each preset time period; A first threshold adjustment module is configured to, if there is a scene range whose corresponding secondary confirmation efficiency rate shows a continuous downward trend within the preset time period, determine the scene range whose corresponding secondary confirmation efficiency rate shows a continuous downward trend within the preset time period as a first target scene range, and increase the alarm threshold of each camera in each first target scene range; The second threshold adjustment module is used to determine the scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period as the second target scene range if there is a scene range whose corresponding secondary confirmation efficiency shows a continuous upward trend within the preset time period, and lower the alarm threshold of each camera in each second target scene range.

8. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to execute the steps of the scene-linked video surveillance method as described in any one of claims 1 to 6.

9. A video surveillance system based on scene linkage, characterized in that: The system includes a monitoring platform and a monitoring terminal, wherein the monitoring terminal includes at least one camera and a robot; The monitoring platform is used to execute the steps of the scene-linkage-based video monitoring method as described in any one of claims 1 to 6.

Images