Scene monitoring method and device, electronic equipment and storage medium

Abstract

The application provides a scene monitoring method and device, electronic equipment and storage medium, and relates to the technical field of intelligent monitoring, which comprises the following steps: comparing current scene information of a target scene with preset scene information configured in advance to obtain a comparison result; determining a monitoring task corresponding to the target scene and a target monitoring component corresponding to the monitoring task according to the comparison result; calling each monitor in the target monitoring component to monitor current environmental information of the target scene according to the monitoring task; and detecting the current environmental information according to the monitoring task to obtain a monitoring result of the target scene. The application realizes adaptive monitoring of a scene according to computing power resources and network bandwidth resources, and further efficiently and accurately monitors the scene.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, and in particular to a scene monitoring method, device, electronic device, and storage medium. Background Technology

[0002] Real-time monitoring of critical scenarios is a crucial prerequisite for ensuring their safety and order. In laboratory settings, for example, laboratories, as original innovation bases, play a vital role in national basic research, technological development, and tackling key technological challenges. Inadequate monitoring and management of laboratories can lead to significant economic and property losses, and in severe cases, even personal injury or death. Therefore, implementing laboratory scenario monitoring is one of the effective means to ensure laboratory safety.

[0003] Traditional laboratory safety management, such as monitoring the placement of anti-static clothing, typically relies on a single monitoring device, such as a camera, to complete the entire monitoring process, including capturing images, performing calculations, and identifying data. However, single monitoring devices usually have limited network bandwidth and computing power, while laboratory scenarios are complex and involve numerous monitoring tasks, making it impossible to achieve efficient and accurate monitoring of the laboratory environment using a single device. Summary of the Invention

[0004] This invention provides a scene monitoring method, device, electronic device, and storage medium to address the shortcomings of existing single monitoring devices in terms of limited monitoring efficiency and accuracy, thereby achieving efficient and accurate scene monitoring.

[0005] This invention provides a scene monitoring method, comprising:

[0006] The current scene information of the target scene is compared with the pre-configured preset scene information to obtain the comparison result;

[0007] Based on the comparison results, the monitoring task corresponding to the target scenario and the target monitoring component corresponding to the monitoring task are determined;

[0008] According to the monitoring task, each monitor in the target monitoring component is invoked to monitor the current environmental information of the target scene;

[0009] Based on the monitoring task, the current environmental information is detected to obtain the monitoring results of the target scene.

[0010] According to a scene monitoring method provided by the present invention, the step of determining the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result includes:

[0011] If the comparison result is true, the monitoring task corresponding to the target scene is determined to include the environmental status monitoring task;

[0012] Based on the environmental status monitoring task, the target monitoring component is determined to include at least one first monitor and at least one second monitor;

[0013] The environmental status monitoring task includes a light monitoring subtask and a person monitoring subtask. The first monitor is used to monitor the light information required by the light monitoring subtask in the current environmental information, and the second monitor is used to monitor the person information required by the person monitoring subtask in the current environmental information.

[0014] According to a scene monitoring method provided by the present invention, the step of invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task includes:

[0015] When the monitoring task includes the environmental status monitoring task, according to the light monitoring sub-task, each first monitor is invoked to monitor the light information of the target scene, and according to the person monitoring sub-task, each second monitor is invoked to monitor the person information of the target scene.

[0016] The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes:

[0017] Based on the lighting monitoring subtask, the lighting information of the target scene is detected to obtain the lighting detection result;

[0018] If, based on the light detection results, it is determined that the current light state in the target scene is on, the person information is detected according to the person monitoring sub-task to obtain the person detection result.

[0019] Based on the person detection results, the monitoring results of the target scene are determined.

[0020] According to a scene monitoring method provided by the present invention, determining the monitoring result of the target scene based on the person detection result includes:

[0021] If, based on the person detection results, it is determined that the target scene contains a person, and the person stays in the target scene for a preset duration, then the monitoring result is determined to be that the current environmental state of the target scene is normal.

[0022] If, based on the character detection results, it is determined that there are no characters in the target scene, the monitoring result indicates that the current environmental state is in an abnormal state.

[0023] According to a scene monitoring method provided by the present invention, the step of determining the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result includes:

[0024] If the comparison result is false, it is determined that the monitoring task includes a security protection monitoring task, and the target monitoring component includes at least one third monitor.

[0025] According to a scene monitoring method provided by the present invention, the step of invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task includes:

[0026] When the monitoring task includes the security protection monitoring task, each third monitor is invoked to monitor the security protection information in the target scenario according to the security protection monitoring task.

[0027] The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes:

[0028] Based on the security protection monitoring task, the security protection information in the target scenario is detected to obtain the security protection detection result;

[0029] If, based on the security protection detection results, it is determined that the target person in the target scene is not wearing special clothing and the target person stays in the target scene for a preset duration, then the monitoring result of the target scene is determined to be that the current security protection status is abnormal.

[0030] If, based on the security protection detection results, it is determined that the target person is wearing the special clothing, then the monitoring result of the target scene is determined to be that the current security protection status is normal.

[0031] According to a scene monitoring method provided by the present invention, the method further includes:

[0032] If the monitoring results determine that the target scene is a normal scene, the timer corresponding to the monitoring task is reset.

[0033] If the target scenario is determined to be an abnormal scenario, the timer is controlled to continue counting.

[0034] If the timer's duration exceeds a time threshold, an alarm is generated based on the monitoring results, and the alarm is pushed out.

[0035] If the alarm message is successfully pushed, reset the timer.

[0036] According to a scene monitoring method provided by the present invention, the step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes:

[0037] Based on the monitoring task and the configuration information of each monitor, load the scene detection algorithm corresponding to each monitor;

[0038] Based on the scene detection algorithm corresponding to each monitor, the current environmental information obtained by each monitor is detected to obtain the monitoring result of the target scene.

[0039] According to a scene monitoring method provided by the present invention, the current scene information includes the current time point, and the preset scene information includes a time flag bit;

[0040] The step of comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result includes:

[0041] Compare the current time point with the time flag bit;

[0042] If the current time point exceeds the time flag bit, the comparison result is determined to be true;

[0043] If the current time point has not exceeded the time flag bit, the comparison result is determined to be false.

[0044] The present invention also provides a scene monitoring device, comprising:

[0045] The comparison unit is used to compare the current scene information of the target scene with the pre-configured preset scene information to obtain the comparison result;

[0046] The determining unit is used to determine the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result.

[0047] The invocation unit is used to invoke each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task.

[0048] The monitoring unit is used to detect the current environmental information according to the monitoring task and obtain the monitoring results of the target scene.

[0049] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the scene monitoring method as described above.

[0050] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the scene monitoring method as described above.

[0051] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the scene monitoring method as described above.

[0052] The scene monitoring method, apparatus, electronic device, and storage medium provided by this invention, through the scene monitoring method provided in this embodiment, compare the current scene information of the target scene with pre-configured preset scene information through an edge server, adaptively divide monitoring tasks and determine target monitoring components based on the comparison results, and utilize the computing power and network bandwidth resources of each monitor in the target monitoring component to monitor the current environmental information of the target scene, and utilize the computing power and network bandwidth resources of the edge server to detect the current environmental information, thereby realizing scene monitoring. By fully combining the computing power and network bandwidth resources of the edge server and the computing power and network bandwidth resources of each monitor in the target monitoring component according to the monitoring tasks, adaptive monitoring of the scene is achieved, avoiding the influence of the limited monitoring efficiency and accuracy of a single monitoring device, and thus achieving efficient and accurate scene monitoring. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0054] Figure 1 This is one of the flowcharts illustrating the scene monitoring method provided by the present invention;

[0055] Figure 2 This is a schematic diagram of the structure of the scene monitoring platform of the scene monitoring method provided by the present invention;

[0056] Figure 3 This is the second flowchart of the scene monitoring method provided by the present invention;

[0057] Figure 4 This is a schematic diagram of the scene monitoring device provided by the present invention;

[0058] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0060] Traditional laboratory safety management, such as monitoring the placement of anti-static clothing, is typically one-off and non-networked. This often involves a single monitoring device, such as a panel machine or other standing device or camera placed at the laboratory entrance, that detects and reports passing personnel only once. This results in poor monitoring effectiveness, rendering laboratory monitoring merely a formality. Furthermore, due to the limitations of the computing power of single monitoring devices, it is difficult to guarantee monitoring effectiveness and efficiency in laboratory monitoring scenarios requiring complex reasoning logic. Customizing monitoring for complex scenarios further increases monitoring costs significantly. Therefore, current technologies cannot efficiently and accurately monitor complex scenarios using a single monitoring device.

[0061] To address the aforementioned issues, this embodiment provides a scene monitoring method and apparatus. By employing a collaborative approach involving a server and multiple monitors, it achieves online, end-to-end scene monitoring. Specifically, the server uses customized scene inference to schedule multiple monitors in real time, analyzes the data collected by the monitors in real time, obtains real-time scene monitoring results, and pushes the monitoring results in real time. This method is simple to maintain and enables efficient and accurate scene monitoring.

[0062] The following is combined Figure 1 The present invention provides a scene monitoring method.

[0063] like Figure 1 The diagram shown is one of the flowcharts of the scene monitoring method provided in this embodiment of the invention. The scene monitoring method provided in this embodiment can be applied to a scene monitoring platform, which includes an edge server, multiple monitors, and other hardware devices for assisting scene monitoring, such as access control and loudspeakers. Specifically, these can be added or removed adaptively according to the scene monitoring requirements.

[0064] Traditional intelligent monitoring frameworks typically rely on the computing power of the cameras themselves for simple feature recognition, failing to effectively monitor complex scenes. Furthermore, traditional NVRs (Network Video Recorders) cannot display alarms based on specific identified information, hindering comprehensive monitoring and resulting in poor monitoring performance. In contrast, the scene monitoring platform in this embodiment adheres to decoupling, facilitating the adaptive addition and removal of components. It is easily scalable and low-cost. Moreover, the platform's computing power is limited by the edge server's computing capacity, not by the computing power of a single monitoring device. Therefore, this platform offers higher efficiency and effectiveness for monitoring complex scenes. For example, the 21T computing power of the edge server can support encoding and decoding of 16 1920x1080 cameras and real-time inference of neural network detection algorithms corresponding to multiple cameras, enabling efficient and accurate scene monitoring.

[0065] like Figure 2 The diagram shows the structure of a scene monitoring platform. The edge server communicates with multiple monitors and other hardware devices. The operation, deployment, and maintenance of this scene monitoring platform are all based on the edge server.

[0066] An edge server includes an algorithm engine, a backend, and a frontend, enabling the edge server to perform efficient and accurate scene monitoring based on the communication and interaction between the algorithm engine, backend, frontend, and hardware devices connected to the edge server.

[0067] The algorithm engine stores various detection algorithms and calls them to execute monitoring tasks during scene monitoring. The backend includes Spring Boot, service software, business logic, and a database, used to store monitoring data, process the monitoring data, schedule hardware devices, and interact with the frontend. The frontend is used for content display and human-computer interaction through various software, including WEB (World Wide Web) web pages, mini-programs, nginx (server software), and VUE (video shooting and enhancement tools). This embodiment does not specifically limit these.

[0068] It should be noted that the edge server can be used as the execution subject of the scene monitoring method provided in this embodiment; the following describes the scene monitoring method in this embodiment in detail with the edge server as the execution subject. The method includes the following steps:

[0069] Step 101: Compare the current scene information of the target scene with the pre-configured preset scene information to obtain the comparison result;

[0070] The target scenario refers to the scenario that needs to be monitored, including but not limited to laboratories, factory workshops, etc. This embodiment does not specifically limit this.

[0071] The current scene information is information used to characterize the current state of the target scene, including time information or environmental information, etc. This embodiment does not specifically limit this.

[0072] The preset scene information is pre-configured by the edge server based on the monitoring requirements of the target scene.

[0073] Optionally, the edge server first collects the current scene information of the target scene and loads the pre-configured preset scene information; it then compares the current scene information with the preset scene information to divide the monitoring tasks according to the comparison results.

[0074] Step 102: Based on the comparison results, determine the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task;

[0075] Different comparison results correspond to different monitoring tasks, and different monitoring tasks correspond to different monitoring components. That is, the environmental information used to implement different monitoring tasks is different, and the monitoring components used to collect different environmental information are different.

[0076] The target monitoring component includes one or more types of monitors. The number of monitors in the target monitoring component and the configuration information of each monitor can be pre-configured according to the monitoring requirements of the target scene, such as configuring 5 monitors. The monitors in the target monitoring component can be cameras or other devices with monitoring functions. The configuration information includes basic information of the monitor, one or more monitoring algorithms, and the mapping relationship between the monitor and the scene detection algorithm. This embodiment does not specifically limit this.

[0077] Monitoring tasks include, but are not limited to, environmental status monitoring tasks or security protection monitoring tasks. The specific tasks can be set according to the monitoring requirements, and this embodiment does not make any specific limitations on them.

[0078] Optionally, after obtaining the comparison results, the monitoring task corresponding to the target scene can be determined based on the correspondence between the comparison results and the monitoring tasks, and the monitor used to implement the environmental information collection function corresponding to the monitoring task can be determined to construct the target monitoring component.

[0079] Step 103: According to the monitoring task, call each monitor in the target monitoring component to monitor the current environmental information of the target scene;

[0080] A monitoring task may include one or more sub-monitoring tasks. The specific number can be set according to the actual monitoring needs. For example, a monitoring task may include a light monitoring sub-task and a personnel monitoring sub-task, or a monitoring task may only include a security protection monitoring sub-task.

[0081] Optionally, after obtaining the monitoring task, the number of subtasks contained in the monitoring task can be determined. If the monitoring task contains only one subtask, the monitors in the target monitoring component can be directly invoked to synchronously monitor the target scene to obtain the current environmental information of the target scene. If the monitoring task contains multiple subtasks, the monitors in the target monitoring component can be directly invoked to synchronously monitor the target scene, obtain the environmental information corresponding to each subtask, and thus obtain the current environmental information; or, the execution order of each subtask can be determined. Based on the execution order of each subtask and the monitoring functions of each monitor in the target monitoring component, the monitoring order of each monitor in the target monitoring component can be determined. Based on the monitoring order, each monitor in the target monitoring component can be invoked sequentially to monitor the target scene, obtain the environmental information corresponding to each subtask, and thus obtain the current environmental information. This embodiment does not specifically limit this.

[0082] Step 104: Based on the monitoring task, detect the current environmental information to obtain the monitoring results of the target scene.

[0083] Optionally, after obtaining the current environmental information, a detection algorithm can be determined according to the monitoring task, and the current environmental information can be detected according to the detection algorithm. Based on the detection results, it can be determined whether the target scene is an abnormal scene or a normal scene, so as to facilitate effective supervision of the target scene.

[0084] The detection algorithm can be determined jointly based on the configuration information of each monitor in the target monitoring component and the monitoring task; or it can be determined directly based on the detection algorithm required by the monitoring task. This embodiment does not specifically limit this.

[0085] Understandably, compared to the traditional model of using a single monitoring device, which suffers from limitations in monitoring efficiency and accuracy, the scene monitoring method provided in this embodiment has its computing power limit and network resource bandwidth jointly determined by the computing power limits of the edge server and the target monitor components, rather than the computing power and network resource bandwidth of a single monitoring device. Therefore, for algorithm customization in complex scenarios, this method has a higher computing power limit and better monitoring effect, thereby achieving efficient and accurate scene monitoring. Furthermore, the entire monitoring process is based on real-time scheduling of multiple monitors and real-time analysis by the edge server, following a decoupling model. This decouples the environmental information required for the monitoring task, facilitating the adaptive selection (addition or deletion) of appropriate monitors to monitor environmental information during the execution of the monitoring task, without the need for additional physical hardware. This approach is characterized by easy scalability, high flexibility, and low cost.

[0086] It should be noted that the above scene monitoring method can be applied to iterative monitoring of the target scene in each cycle, so as to monitor the scene in real time and efficiently. The scene monitoring steps for each cycle can be found in steps 101 to 104, which will not be repeated here.

[0087] The scene monitoring method provided in this embodiment compares the current scene information of the target scene with pre-configured preset scene information through an edge server. Based on the comparison results, it adaptively divides monitoring tasks and determines target monitoring components. It then utilizes the computing power and network bandwidth resources of each monitor in the target monitoring component to monitor the current environmental information of the target scene, and uses the computing power and network bandwidth resources of the edge server to detect the current environmental information, thereby achieving scene monitoring. By fully combining the computing power and network bandwidth resources of the edge server and each monitor in the target monitoring component according to the monitoring tasks, the method adaptively monitors the scene, avoiding the limitations of the monitoring efficiency and accuracy of a single monitoring device, and thus achieving efficient and accurate scene monitoring.

[0088] In some embodiments, step 102, which involves determining the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result, further includes:

[0089] If the comparison result is true, the monitoring task corresponding to the target scene is determined to include the environmental status monitoring task;

[0090] Based on the environmental status monitoring task, the target monitoring component is determined to include at least one first monitor and at least one second monitor;

[0091] The environmental status monitoring task includes a light monitoring subtask and a person monitoring subtask. The first monitor is used to monitor the light information required by the light monitoring subtask in the current environmental information, and the second monitor is used to monitor the person information required by the person monitoring subtask in the current environmental information.

[0092] The comparison results are divided into true or false, and the monitoring tasks are divided and the monitoring components are determined based on the true or false results.

[0093] The environmental status monitoring task is used to detect the environmental status of the target scene, including but not limited to the lighting monitoring subtask and the personnel monitoring subtask, which are determined according to the monitoring requirements of the target scene. The lighting monitoring subtask is used to detect the lighting status in the target environment, and the personnel monitoring subtask is used to detect the status of personnel in the target environment.

[0094] Optionally, after obtaining the comparison result, it is necessary to determine whether the comparison result is true or false; if the comparison result is true, it is determined that the monitoring task corresponding to the target scene includes the task for detecting the environmental state of the target scene, i.e., the environmental state monitoring task.

[0095] Therefore, the target monitoring component corresponding to the environmental status monitoring task is determined to be a monitoring component used to monitor the environmental information required for the environmental status monitoring task. This component includes at least one first monitor for monitoring the lighting information required for the lighting monitoring sub-task, and at least one second monitor for monitoring the personnel information required for the personnel monitoring sub-task. The specific number of the first and second monitors can be set according to actual needs, such as 1 first monitor and 4 second monitors.

[0096] In this embodiment, if the comparison result is true, it can be determined that the monitoring task includes the environmental status monitoring task. By combining the bandwidth resources and computing power inference resources of at least one first monitor and at least one second monitor, the monitoring of the environmental information required for the environmental status monitoring task can be achieved quickly and accurately, thereby achieving efficient and accurate monitoring of the target scene.

[0097] In some embodiments, step 103, which involves invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task, further includes:

[0098] When the monitoring task includes the environmental status monitoring task, according to the light monitoring sub-task, each first monitor is invoked to monitor the light information of the target scene, and according to the person monitoring sub-task, each second monitor is invoked to monitor the person information of the target scene.

[0099] Optionally, when the monitoring task includes the environmental status monitoring task, the monitoring order of the first monitor and the monitoring order of the second monitor can be determined according to the execution order of the light monitoring sub-task and the execution order of the person monitoring sub-task. That is, the monitoring order of the first monitor is consistent with the execution order of the light monitoring sub-task, and the monitoring order of the second monitor is consistent with the execution order of the person monitoring sub-task.

[0100] Then, according to the monitoring order, the lighting information of the target scene monitored by each first monitor and the personnel information of the target scene monitored by each second monitor are called in turn.

[0101] For example, if the execution order of the lighting monitoring subtask is before the person monitoring subtask, then the monitoring order of the first monitor is before the second monitor. Accordingly, each first monitor is called first to monitor the lighting information of the target scene, and then each second monitor is called to monitor the person information of the target scene.

[0102] Step 104, which involves detecting the current environmental information based on the monitoring task to obtain the monitoring result of the target scene, further includes:

[0103] Based on the lighting monitoring subtask, the lighting information of the target scene is detected to obtain the lighting detection result;

[0104] If, based on the light detection results, it is determined that the current light state in the target scene is on, the person information is detected according to the person monitoring sub-task to obtain the person detection result.

[0105] Based on the person detection results, the monitoring results of the target scene are determined.

[0106] Optionally, upon obtaining lighting information, a lighting detection algorithm can be determined first based on the configuration information of the lighting monitoring subtask and the first monitor. This algorithm is then used to detect the lighting information of the target scene to determine whether the current lighting state of the target scene is on. If the lighting is off, the process returns to step 101 and iteratively monitors the target scene.

[0107] When the lights are on, the system acquires the information of people in the target scene monitored by the second monitor. Based on the person monitoring sub-task and the configuration information of the second monitor, it determines the person detection algorithm. The system then uses the person detection algorithm to detect the person information in the target scene to determine whether the person information in the target scene meets the conditions. Based on the determination result, the system acquires the monitoring result of the target scene.

[0108] In this embodiment, based on the environmental status monitoring task, the first and second monitors are adaptively invoked to collect lighting and human information of the target scene for lighting detection and human detection. Without requiring additional hardware (monitor) version updates, environmental status monitoring of the target scene can be flexibly and cost-effectively achieved, enabling efficient and accurate environmental status monitoring of the target scene.

[0109] In some embodiments, determining the monitoring result of the target scene based on the person detection result includes:

[0110] If, based on the person detection results, it is determined that the target scene contains a person, and the person stays in the target scene for a preset duration, then the monitoring result is determined to be that the current environmental state of the target scene is normal.

[0111] If, based on the character detection results, it is determined that there are no characters in the target scene, the monitoring result indicates that the current environmental state is in an abnormal state.

[0112] Among them, the person detection is used to detect whether there are people in the target scene. If there are people, it is necessary to determine the duration of their stay in order to determine whether there are people working normally in the target scene when the lights are on.

[0113] The preset duration can be set according to monitoring needs to limit the effective tracking time of a person.

[0114] Optionally, after obtaining the detection results, it is necessary to determine whether the target scene contains people based on the detection results. If no people are contained, the monitoring result is determined to be that the current environment is in an abnormal state, that is, the target scene is determined to be an abnormal scene.

[0115] If a person is present in the target scene, the detection results will further determine if the person's dwell time in the target scene reaches a preset duration, indicating that the person is a valid person (i.e., a normal working person) in the target scene. Further, if a normal working person is present in the target scene while the lights are on, the target scene is determined to be in a normal state. Otherwise, the person is considered an invalid person in the target scene (i.e., an invalid person caused by monitoring errors or non-working invalid persons). Further, if a normal working person is not present in the target scene while the lights are on, the target scene is determined to be in an abnormal state, and an alarm will be triggered to allow management personnel to promptly turn off the lights in the target scene to avoid resource waste and impact on laboratory safety.

[0116] In some embodiments, step 102, which involves determining the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result, further includes:

[0117] If the comparison result is false, it is determined that the monitoring task includes a security protection monitoring task, and the target monitoring component includes at least one third monitor.

[0118] Optionally, after obtaining the comparison result, it is necessary to determine whether the comparison result is true or false; if the comparison result is false, the monitoring task corresponding to the target scene is determined to include the task used to detect the security protection status of the target scene, i.e., the security protection monitoring task.

[0119] Therefore, the target monitoring component corresponding to the security protection monitoring task is determined to be a monitoring component used to monitor the security protection information required for the security protection monitoring task. This component includes at least one third monitor. The at least one third monitor may include a first monitor and / or a second monitor, or may include other monitors besides the first and second monitors. Specifically, it can be determined according to the configuration information of the monitors. This embodiment does not make a specific limitation, such as one first monitor and four second monitors.

[0120] It should be noted that when at least one third monitor includes a first monitor and a second monitor, the first monitor must be equipped with both light information monitoring and security protection information monitoring functions, and the second monitor must be equipped with both personnel information monitoring and security protection information monitoring functions.

[0121] In this embodiment, if the comparison result is false, it can be determined that the monitoring task includes a security protection monitoring task. By combining the bandwidth resources and computing power inference resources of at least one third monitor, the monitoring of the environmental information required for the security protection monitoring task can be carried out quickly and accurately, thereby achieving efficient and accurate monitoring of the target scene.

[0122] In some embodiments, step 103, which involves calling each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task, further includes:

[0123] When the monitoring task includes the security protection monitoring task, each third monitor is invoked to monitor the security protection information in the target scenario according to the security protection monitoring task.

[0124] Optionally, if the monitoring task includes security protection monitoring, the security protection information of the target scene can be called simultaneously from each third monitor.

[0125] Safety protection information refers to information that ensures the safe operation of the target environment, including information such as the protective equipment worn by the target personnel. This embodiment does not specifically limit this.

[0126] Step 104, which involves detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene, further includes:

[0127] Based on the security protection monitoring task, the security protection information in the target scenario is detected to obtain the security protection detection result;

[0128] Based on the security protection monitoring task, the security protection information in the target scenario is detected to obtain the security protection detection result;

[0129] If, based on the security protection detection results, it is determined that the target person in the target scene is not wearing special clothing and the target person stays in the target scene for a preset duration, then the monitoring result of the target scene is determined to be that the current security protection status is abnormal.

[0130] If, based on the security protection detection results, it is determined that the target person is wearing the special clothing, then the monitoring result of the target scene is determined to be that the current security protection status is normal.

[0131] The target person refers to someone who needs to wear special clothing within the target scene. Special clothing includes, but is not limited to, antistatic clothing and fireproof clothing; this embodiment does not specifically limit these categories.

[0132] Optionally, upon obtaining security protection information, a security protection detection algorithm can be determined based on the security protection monitoring task and the configuration information of the third monitor. The security protection information of the target scene can then be detected based on the security protection detection algorithm to determine whether the target person in the target scene is wearing special clothing and the duration of the target person's stay in the target scene, thereby determining whether the current security protection status is safe.

[0133] If the target person in the target scenario is not wearing special clothing and the target person stays in the target scenario for a preset duration, it indicates that the current security protection status is abnormal. At this time, it is determined that the target scenario is in an abnormal state, and an alarm is issued to remind the target person to put on special clothing in time to avoid affecting laboratory safety.

[0134] If the target person in the target scene is wearing special clothing and the target person stays in the target scene for a preset time, it indicates that the current security protection status is normal, and we can continue to return to step 101 to iteratively monitor the target scene.

[0135] In this embodiment, a third monitor is adaptively invoked to collect security protection information of the target scene according to the security protection monitoring task, so as to perform security protection information detection. Without requiring additional hardware version updates, the security protection status monitoring of the target scene can be flexibly achieved at low cost, so as to achieve efficient and accurate monitoring of the target scene.

[0136] In some embodiments, the method further includes:

[0137] If the monitoring results determine that the target scene is a normal scene, the timer corresponding to the monitoring task is reset.

[0138] If the target scenario is determined to be an abnormal scenario, the timer is controlled to continue counting.

[0139] If the timer's duration exceeds a time threshold, an alarm is generated based on the monitoring results, and the alarm is pushed out.

[0140] If the alarm message is successfully pushed, reset the timer.

[0141] The timer provides a time reference for alarms in abnormal scenarios.

[0142] Optionally, based on the monitoring results, it can be determined whether the target scenario is a normal scenario, and based on the determination result, the control operations and alarm operations to be performed on the timer can be determined.

[0143] For example, when performing an environmental status monitoring task, if the current environmental status of the target scene is determined to be normal based on the monitoring results, then the target scene is determined to be a normal scene, the timer's timing can be initialized to 0, and the process can return to step 101 to iteratively monitor the target scene; if the current environmental status of the target scene is determined to be abnormal based on the monitoring results, then the target scene is determined to be an abnormal scene, the timer can be controlled to continue timing, and it can be determined whether the timer's timing is greater than a time threshold. If the timer is greater than the time threshold, an alarm message is generated based on the monitoring results; the alarm message includes at least the monitoring results and the task information of the monitoring task, etc., which are not specifically limited in this embodiment.

[0144] The alarm information is synchronously pushed to the backend. After receiving the alarm information pushed by the inference engine, the backend can forward the results or perform operations according to different monitoring tasks. During this process, the pushed alarm information is stored in the permanent storage space of the edge server and added to the frontend call of the edge server. At the same time, the alarm information can also be synchronized to the cloud server. In addition, the backend can call physical resources in the target scene, such as speakers and loudspeakers, in the form of scripts to trigger alarms based on the alarm information, so that managers can promptly handle abnormal situations in the target scene, facilitate the timely restoration of the target scene to a normal state, and improve the security of the target scene.

[0145] In addition, once it is confirmed that the alarm information has been successfully pushed to the backend, the timer is filtered at the same time, that is, the timer is reset and the timer continues to be timed a second time to prevent the alarm information from being pushed to the backend continuously, which would interfere with the staff in the target scene and affect the user experience.

[0146] In some embodiments, step 104, which involves detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene, includes:

[0147] Based on the monitoring task and the configuration information of each monitor, load the scene detection algorithm corresponding to each monitor;

[0148] Based on the scene detection algorithm corresponding to each monitor, the current environmental information obtained by each monitor is detected to obtain the monitoring result of the target scene.

[0149] It should be noted that before scene monitoring, the edge server pre-configures the configuration information of each camera according to the monitoring requirements of the target scene. This configuration information includes, but is not limited to, the basic information of the monitor, one or more monitoring algorithms, and the mapping relationship between the monitor and the scene detection algorithm. This embodiment does not specifically limit this. For example, configuring the first monitor includes a light monitoring algorithm and a security protection monitoring algorithm, as well as the mapping relationship between the first monitor and the light detection algorithm, and the mapping relationship between the first monitor and the security protection detection algorithm, etc.

[0150] Optionally, when performing step 104, it is necessary to first load the scene detection algorithms corresponding to each monitor required to implement the monitoring task into the algorithm database according to the mapping relationship in the monitoring task and the configuration information of each monitor.

[0151] Then, based on the scene detection algorithm corresponding to each monitor, the current environmental information obtained by each monitor is detected to obtain the monitoring results of the target scene.

[0152] For example, when the monitoring task includes a light detection subtask, the scene detection algorithm corresponding to the first monitor is a light detection algorithm. The light information is detected based on the light detection algorithm to obtain the light detection result, so as to obtain the monitoring result of the target scene based on the light detection result.

[0153] Compared to traditional scene monitoring methods, this embodiment uses an edge server to achieve real-time scheduling of multiple monitors, and performs real-time analysis and push of the current environmental information monitored by each monitor based on the detection algorithm corresponding to each monitor. This simplifies operation and maintenance, thereby improving the efficiency, completeness and real-time performance of scene monitoring.

[0154] In some embodiments, the current scene information includes the current time point, and the preset scene information includes a time flag bit;

[0155] The step of comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result includes:

[0156] Compare the current time point with the time flag bit;

[0157] If the current time point exceeds the time flag bit, the comparison result is determined to be true;

[0158] If the current time point has not exceeded the time flag bit, the comparison result is determined to be false.

[0159] The time stamp can be pre-configured according to the monitoring requirements of the target scene to divide tasks and make reasonable use of the bandwidth and computing power resources of each monitor. This enables real-time scheduling, analysis and inference of different algorithm logics in the laboratory at different time periods. In addition, a highly reusable solution is provided by using module decoupling, which facilitates the development, deployment and operation of customized edge scenarios and improves the overall computing power of the entire target scene monitoring. By allocating and calling resources at the application layer of the edge server, the difficulties of developing, deploying and calling applications for single-camera algorithms are solved, and efficient and accurate scene monitoring is achieved.

[0160] Optionally, the current time of the target scene can be obtained in real time, and it can be determined whether the current time exceeds the time flag. If it exceeds the time flag, the comparison result is determined to be true, that is, the current time is night and an environmental status monitoring task needs to be performed; otherwise, the comparison result is determined to be false, that is, the current time is day and a security protection monitoring task needs to be performed.

[0161] By comparing the current time of the target scene with the time flag, tasks can be rationally divided to enable security protection monitoring tasks during the day and environmental status monitoring tasks at night, thereby increasing the monitoring range and integrity of the target scene and ensuring more effective scene monitoring.

[0162] The following describes the scene monitoring method in this embodiment using specific examples.

[0163] like Figure 3 The diagram shown is a second flowchart of the scene monitoring method provided in this embodiment. The method includes:

[0164] Step 301: Initialize the configuration of the monitoring algorithms and scene detection algorithms of each monitor, the mapping relationships between the monitors, and the initialization of interface information, such as time flags. For example, there are 5 monitors: one first monitor (monitor 1) and four second monitors (monitor 2, monitor 3, monitor 4, and monitor 5). The configuration of the first monitor includes the light monitoring algorithm, the mapping relationship between the first monitor and the light detection algorithm, and the security monitoring algorithm, the mapping relationship between the first monitor and the security detection algorithm. The configuration of the second monitors includes the person monitoring algorithm, the mapping relationship between the second monitor and the person detection algorithm, and the security monitoring algorithm, the mapping relationship between the second monitor and the security detection algorithm.

[0165] Step 302: Enter polling mode and compare the current time of the target scene with the pre-configured time flag to determine whether the current time of the target scene exceeds the time flag. If it exceeds the time flag, the comparison result is determined to be true; otherwise, it is false. This allows for the reasonable division of monitoring tasks, enabling security protection monitoring tasks to be performed during the day and environmental status monitoring tasks to be performed at night.

[0166] Step 303: Execute the corresponding monitoring task;

[0167] If the comparison result is true, perform the environmental status monitoring task. The specific steps are as follows:

[0168] Polling is used to detect lights based on the light information collected by monitor 1.

[0169] It should be noted that when monitor 1 is a night vision camera, saturation in the color model (HSV) is used for light detection; when monitor 1 is a non-night vision camera, average brightness is used for light detection.

[0170] If the lighting status of the target scene is determined to be on based on the light detection results, the system collects information about people based on monitors 2, 3, 4 and 5. The system uses the person detection algorithm corresponding to each of the second monitors in monitors 2, 3, 4 and 5, such as the general YOLOX model, to detect people based on the collected information. This determines whether there are people in the target scene. When people are detected in the target scene and it is determined that the time the person stays in the target scene reaches the preset time, that is, the minimum tracking time of the person is met, it indicates that there are still people working normally in the target scene. No alarm is required. After refreshing and resetting the timer, the system returns to step 302 to continue the next round of scene monitoring.

[0171] If the target scene does not contain any people, the timer continues to count down. If the timer's countdown does not meet the condition of exceeding the time threshold, the process returns to step 302 to continue the next round of scene monitoring. If the timer's countdown exceeds the time threshold, an alarm message will be generated based on the monitoring results and pushed to the backend. At the same time, the timer will be filtered to perform a second countdown, preventing the continuous push of alarm messages from interfering with the staff.

[0172] If the comparison result is false, perform a security monitoring task. The specific steps are as follows:

[0173] The controllers 1, 2, 3, 4 and 5 collect safety protection information in the target environment and perform safety protection detection based on the corresponding safety protection detection algorithm of each controller, such as the YOLOX model for special clothing. If, based on the detection results, it is determined that the target person in the target scene is not wearing antistatic clothing and the target person stays in the target scene for a preset time, an alarm message is generated and pushed to the backend before returning to step 302.

[0174] In summary, the scene monitoring method provided in this embodiment achieves online scene monitoring based on a joint monitoring framework of edge servers and multiple monitors. Since the upper limit of computing power under this framework is determined by the computing power limit of the edge servers (rather than the computing power of a single monitor), it offers a higher upper limit and better inference performance for algorithm customization in complex scenarios compared to traditional methods. Furthermore, the operation, deployment, and maintenance under this framework are all based on the application layer of the edge servers, requiring no hardware updates, making deployment more convenient and efficient. It also enables adaptive and efficient switching between multiple monitoring tasks; that is, based on current environmental information, security protection tasks can be performed during the day to achieve security protection. When an abnormal status occurs, alarm images are uploaded and voice announcements are made. At night, human and light detection is employed to enable lighting control or alarms even when no one is present. Furthermore, without affecting normal business operations, additional algorithm modules can be added to existing monitors in the target scenario (without purchasing panel cameras or standing cameras) to achieve adaptive and efficient switching between multiple monitoring tasks, effectively reducing maintenance costs. Moreover, adhering to a decoupled model between the front-end, back-end, and algorithm engine, the algorithm's inference, push, and display results are isolated, facilitating the addition and modification of various data processing modules and push display modules, thus achieving efficient, accurate, and comprehensive scene monitoring.

[0175] The scene monitoring device provided by the present invention is described below. The scene monitoring device described below and the scene monitoring method described above can be referred to in correspondence.

[0176] like Figure 4 As shown, this embodiment provides a scene monitoring device, which includes:

[0177] The comparison unit 401 is used to compare the current scene information of the target scene with the pre-configured preset scene information to obtain the comparison result;

[0178] The determining unit 402 is used to determine the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result;

[0179] The calling unit 403 is used to call each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task;

[0180] The monitoring unit 404 is used to detect the current environmental information according to the monitoring task and obtain the monitoring result of the target scene.

[0181] The scene monitoring device provided in this embodiment compares the current scene information of the target scene with pre-configured preset scene information through an edge server. Based on the comparison results, it adaptively divides monitoring tasks and determines target monitoring components. It then utilizes the computing power and network bandwidth resources of each monitor in the target monitoring component to monitor the current environmental information of the target scene, and uses the computing power and network bandwidth resources of the edge server to detect the current environmental information, thereby achieving scene monitoring. By fully combining the computing power and network bandwidth resources of the edge server and each monitor in the target monitoring component according to the monitoring tasks, it adaptively monitors the scene, avoiding the limitations of monitoring efficiency and accuracy of a single monitoring device, and thus achieving efficient and accurate scene monitoring.

[0182] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include a processor 501, a communication interface 502, a memory 503, and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other via the communication bus 504. The processor 501 can call logical instructions in the memory 503 to execute a scene monitoring method. This method includes: comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result; determining, based on the comparison result, a monitoring task corresponding to the target scene and a target monitoring component corresponding to the monitoring task; calling each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task; and detecting the current environmental information according to the monitoring task to obtain a monitoring result for the target scene.

[0183] Furthermore, the logical instructions in the aforementioned memory 503 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0184] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the scene monitoring method provided by the above methods. The method includes: comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result; determining the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result; calling each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task; and detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene.

[0185] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the scene monitoring method provided by the above methods. The method includes: comparing the current scene information of a target scene with pre-configured preset scene information to obtain a comparison result; determining, based on the comparison result, a monitoring task corresponding to the target scene and a target monitoring component corresponding to the monitoring task; according to the monitoring task, invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene; and according to the monitoring task, detecting the current environmental information to obtain a monitoring result of the target scene.

[0186] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0187] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A scene monitoring method, characterized in that, include: The current scene information of the target scene is compared with the pre-configured preset scene information to obtain the comparison result; Based on the comparison results, the monitoring task corresponding to the target scenario and the target monitoring component corresponding to the monitoring task are determined; According to the monitoring task, each monitor in the target monitoring component is invoked to monitor the current environmental information of the target scene; Based on the monitoring task, the current environmental information is detected to obtain the monitoring results of the target scene; The steps for determining the monitoring task corresponding to the target scenario include: If the comparison result is true, the monitoring task corresponding to the target scene is determined to include an environmental status monitoring task; if the comparison result is false, the monitoring task corresponding to the target scene is determined to include a security protection monitoring task. The environmental status monitoring task includes a light monitoring sub-task and a personnel monitoring sub-task; The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes: Based on the lighting monitoring subtask, the lighting information of the target scene is detected to obtain the lighting detection result; If, based on the light detection results, it is determined that the current light state in the target scene is on, then, based on the person monitoring sub-task, the person information required by the person monitoring sub-task in the current environment information is detected to obtain the person detection result. Based on the person detection results, determine the monitoring results of the target scene; The step of comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result includes: Compare the current time point in the current scene information with the terminal time flag bit of the preset scene information; If the current time point exceeds the time flag bit, the comparison result is determined to be true; If the current time point has not exceeded the time flag bit, the comparison result is determined to be false.

2. The scene monitoring method according to claim 1, characterized in that, If the comparison result is true, the steps for determining the target monitoring component corresponding to the monitoring task include: Based on the environmental status monitoring task, the target monitoring component is determined to include at least one first monitor and at least one second monitor; The first monitor is used to monitor the lighting information required by the lighting monitoring sub-task in the current environment information, and the second monitor is used to monitor the person information required by the person monitoring sub-task in the current environment information.

3. The scene monitoring method according to claim 2, characterized in that, The step of invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task includes: When the monitoring task includes the environmental status monitoring task, according to the light monitoring sub-task, each first monitor is invoked to monitor the light information of the target scene, and according to the person monitoring sub-task, each second monitor is invoked to monitor the person information of the target scene.

4. The scene monitoring method according to claim 3, characterized in that, The step of determining the monitoring results of the target scene based on the person detection results includes: If, based on the person detection results, it is determined that the target scene contains a person, and the person stays in the target scene for a preset duration, then the monitoring result is determined to be that the current environmental state of the target scene is normal. If, based on the character detection results, it is determined that there are no characters in the target scene, the monitoring result indicates that the current environmental state is in an abnormal state.

5. The scene monitoring method according to claim 1, characterized in that, If the comparison result is false, the steps for determining the target monitoring component corresponding to the monitoring task include: Based on the security protection and monitoring task, the target monitoring component is determined to include at least one third monitor.

6. The scene monitoring method according to claim 5, characterized in that, The step of invoking each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task includes: When the monitoring task includes the security protection monitoring task, each third monitor is invoked to monitor the security protection information in the target scenario according to the security protection monitoring task. The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes: Based on the security protection monitoring task, the security protection information in the target scenario is detected to obtain the security protection detection result; If, based on the security protection detection results, it is determined that the target person in the target scene is not wearing special clothing and the target person stays in the target scene for a preset duration, then the monitoring result of the target scene is determined to be that the current security protection status is abnormal. If, based on the security protection detection results, it is determined that the target person is wearing the special clothing, then the monitoring result of the target scene is determined to be that the current security protection status is normal.

7. The scene monitoring method according to any one of claims 1-6, characterized in that, The method further includes: If the monitoring results determine that the target scene is a normal scene, the timer corresponding to the monitoring task is reset. If the target scenario is determined to be an abnormal scenario, the timer is controlled to continue counting. If the timer's duration exceeds a time threshold, an alarm is generated based on the monitoring results, and the alarm is pushed out. If the alarm message is successfully pushed, reset the timer.

8. The scene monitoring method according to any one of claims 1-6, characterized in that, The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes: Based on the monitoring task and the configuration information of each monitor, load the scene detection algorithm corresponding to each monitor; Based on the scene detection algorithm corresponding to each monitor, the current environmental information obtained by each monitor is detected to obtain the monitoring result of the target scene.

9. A scene monitoring device, characterized in that, include: The comparison unit is used to compare the current scene information of the target scene with the pre-configured preset scene information to obtain the comparison result; The determining unit is used to determine the monitoring task corresponding to the target scene and the target monitoring component corresponding to the monitoring task based on the comparison result. The invocation unit is used to invoke each monitor in the target monitoring component to monitor the current environmental information of the target scene according to the monitoring task. The monitoring unit is used to detect the current environmental information according to the monitoring task and obtain the monitoring results of the target scene; The steps for determining the monitoring task corresponding to the target scenario include: If the comparison result is true, the monitoring task corresponding to the target scene is determined to include an environmental status monitoring task; if the comparison result is false, the monitoring task corresponding to the target scene is determined to include a security protection monitoring task. The environmental status monitoring task includes a light monitoring sub-task and a personnel monitoring sub-task; The step of detecting the current environmental information according to the monitoring task to obtain the monitoring result of the target scene includes: Based on the lighting monitoring subtask, the lighting information of the target scene is detected to obtain the lighting detection result; If, based on the light detection results, it is determined that the current light state in the target scene is on, then, based on the person monitoring sub-task, the person information required by the person monitoring sub-task in the current environment information is detected to obtain the person detection result. Based on the person detection results, determine the monitoring results of the target scene; The step of comparing the current scene information of the target scene with pre-configured preset scene information to obtain a comparison result includes: Compare the current time point in the current scene information with the terminal time flag bit of the preset scene information; If the current time point exceeds the time flag bit, the comparison result is determined to be true; If the current time point has not exceeded the time flag bit, the comparison result is determined to be false.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the scene monitoring method as described in any one of claims 1 to 8.

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

Images