An automatic control method and system for fog scenes and a computer readable storage medium
By identifying personnel in real time and generating structured data in the fog system, and dynamically controlling the spray zones, the problem of inflexible control and inconvenient management of existing fog systems is solved, achieving efficient and precise spray control and remote management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHENGZHOU GUORUN TECH CO LTD
- Filing Date
- 2026-03-30
- Publication Date
- 2026-06-12
AI Technical Summary
The existing fog system has a relatively simple control method, which cannot be flexibly adjusted according to the actual working conditions on site. This results in untimely spray control, poor area matching, serious waste of resources, and a lack of remote real-time viewing and management capabilities.
By acquiring video images of the fog scene, using a preset detection model to identify people and generate structured data, dynamically controlling the spray zones, and combining the GB28181 protocol to achieve remote real-time viewing.
It improves the accuracy and flexibility of spray control, reduces resource waste, enhances remote management capabilities, and improves user experience and system efficiency.
Smart Images
Figure CN122200552A_ABST
Abstract
Description
Technical Field
[0001] This invention pertains to the fog scene industry, specifically relating to an automatic control method, system, and computer-readable storage medium for fog scenes. Background Technology
[0002] Fog systems, which combine multiple functions such as landscape creation, cooling and humidification, and air purification, have been widely used in various scenarios such as cultural and tourism fog scenes and commercial display fog effects, becoming an important carrier for enhancing the scene experience and enriching the landscape layers.
[0003] Currently, the control methods of existing fog systems are relatively traditional, mostly adopting three methods: manual control, timed control, or fixed logic control. Their control functions are relatively simple, usually only able to realize the start and stop of the whole system or simple on / off operation of preset fixed areas, and cannot be flexibly adjusted according to the actual working conditions on site. Summary of the Invention
[0004] The purpose of this invention is to provide an automatic control method, system, and computer-readable storage medium for fog scenes, so as to solve the problem of inflexibility in manual control in the prior art.
[0005] To solve the above problems, the present invention provides an automatic control method for fog scenes, comprising: The original video images of a preset area of the fog scene are acquired, and the corresponding target image frames obtained after preprocessing each frame of the original video images are divided into multiple recognition areas; Based on a preset detection model, target personnel are identified in all recognition regions of the target image frame, and the recognition results are output. The recognition results include the target category, the confidence score and bounding box coordinates for each target category. The recognition results are processed based on the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters; the recognition results include target personnel; Based on the structured data and the pre-set control mode, corresponding control instructions are generated, and the execution devices of the corresponding spray zone are controlled according to the control instructions.
[0006] In some embodiments, the target image frame is obtained by performing format conversion, resolution scaling, and letterboxing on each frame of the original video image.
[0007] In some embodiments, processing the recognition result based on parameter information recorded in the preprocessing process to obtain structured data includes: Based on the scaling ratio and padding parameters recorded during the Letterbox processing stage, all bounding box coordinates are reverse-calculated to the original video image coordinates and cropped to the effective range. Based on the target category and target confidence level in the identification results, valid target personnel are selected and the number of target personnel is counted. The center point coordinates of the target person are determined based on the bounding box coordinates of the target person. The target personnel, the number of target personnel, the confidence level of the target personnel, and the coordinates of the center point of the target personnel are encapsulated into structured data.
[0008] In some embodiments, generating corresponding control commands based on the structured data and a pre-set control mode includes: The identification area to which the target person belongs is determined based on the center point coordinates of the target person; Determine the regional status results for the same identification area based on the identification areas to which all target personnel belong; Based on the status results of all regions and the preset control mode, corresponding control commands are generated.
[0009] In some embodiments, the preset control mode includes whether to enable the follow spray mode; If the follow spray mode is enabled, when the area status result of a certain identification area is "person", the corresponding spray zone of the identification area will be activated. If the follow spray mode is not enabled, when the area status result of any identified area is "occupied", control to activate all spray zones.
[0010] In some embodiments, when the number of target personnel in the area status result exceeds a preset number and the stay duration of the target personnel exceeds a preset duration, a control command to activate is triggered.
[0011] To solve the above problems, the present invention relates to an automatic control system for fog scenes, comprising a data acquisition module, a recognition module, a processing module, and a control module; The acquisition module is used to acquire the original video images of a preset area of the fog scene, and divide the corresponding target image frame obtained after preprocessing each frame of the original video image into multiple recognition areas; The recognition module is used to identify target persons in all recognition regions of the target image frame based on a preset detection model, and output the recognition results; the recognition results include the target category, the confidence score and bounding box coordinates corresponding to each target category; The processing module is used to process the recognition results according to the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters; The control module is used to generate corresponding control commands based on the structured data and the pre-set control mode, and to control the execution equipment of the corresponding spray zone according to the control commands.
[0012] In some embodiments, a display module is also included. The display module of the terminal platform is wirelessly connected to the acquisition module. The acquisition module transmits real-time video images of the fog scene to the terminal platform based on the GB28181 protocol and displays the video images through the display module.
[0013] In some embodiments, the actuating device includes a spray host, a solenoid valve, a zone valve, a relay pump control unit, or other switching actuators.
[0014] To address the aforementioned problems, the present invention relates to a computer-readable storage medium storing a computer program, which, when executed on a processor, implements the aforementioned automatic control method for fog scenes.
[0015] The beneficial effects of this invention are as follows: This invention mechanically divides video images into different recognition regions, and obtains identification results based on the recognition processing of all recognition regions. The recognition results are then processed to obtain the number and location of target personnel in a foggy scene in real time. Compared with traditional infrared sensing or timed control methods, this invention has higher recognition accuracy and more complete control data. Furthermore, this invention can also achieve dynamic linkage control of spray zones and remote visual viewing as the location of target personnel changes. Attached Figure Description
[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below: Figure 1 This is a flowchart illustrating the automatic control method for fog scenes in an embodiment of the present invention; Figure 2 This is a schematic diagram of the process for obtaining structured data in an embodiment of the present invention; Figure 3 This is a schematic diagram of the automatic control system for fog scenes in an embodiment of the present invention. Detailed Implementation
[0017] To make the technical objectives, technical solutions, and beneficial effects of the present invention clearer, the technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention; that is, the described embodiments are merely some embodiments of the present invention, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0018] In the following text, the terms "comprising," "having," and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more combinations thereof. Furthermore, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.
[0019] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.
[0020] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.
[0021] The shortcomings of traditional manual control methods are particularly prominent in scenarios with high requirements for user experience and energy consumption control, such as cultural tourism fog scenes, interactive landscape fog forests, and commercial display fog effects. On the one hand, problems such as continuous spraying in unoccupied areas, untimely spraying response in occupied areas, and mismatch between spray coverage and personnel activity areas are prone to occur. This not only seriously affects the user experience and fails to meet the interactive landscape needs of "on-demand fog supply," but also causes a large waste of water and electricity, which is inconsistent with the development trend of energy conservation and environmental protection. On the other hand, existing fog scene systems usually lack a unified network push mechanism and remote real-time viewing means. Managers cannot remotely view the operating status, equipment condition, and fog effect of the spray site in real time through an App or Web interface, which is not conducive to efficient operation and maintenance and timely management of the system, increases manual management costs, and makes it difficult to quickly detect and deal with abnormal problems that occur during system operation. Specifically, existing control methods lack the ability to perceive and respond to on-site personnel distribution, dynamic changes in personnel numbers, and real-time personnel locations. They cannot dynamically adapt and control the spray area, spray duration, and fog volume based on the above-mentioned personnel-related parameters, resulting in a serious lack of control flexibility and accuracy.
[0022] With the rapid development of IoT and intelligent control technologies, and the increasing demands for intelligent, precise, and energy-efficient fog systems in cultural tourism, commerce, and other scenarios, traditional manual control methods can no longer meet practical application needs. Therefore, this invention addresses the problems of low intelligence, untimely control response, crude spray area control, insufficient human-fog interaction, and inadequate on-site video networking and remote viewing capabilities in existing fog system control systems. It proposes the development of a fog system capable of dynamically controlling the spray area based on on-site personnel conditions and possessing remote real-time monitoring capabilities.
[0023] The automatic control method for this fog scene will be described below with reference to some specific embodiments. For example... Figure 1 As shown, the automatic control method for this fog scene includes: S100: Acquire the original video image of the preset area of the fog scene, and divide the corresponding target image frame obtained after preprocessing each frame of the original video image into multiple recognition areas.
[0024] In this embodiment of the invention, to ensure efficient response, the video image is input in real time; as another implementation, it can also be a sequence of images at certain time intervals to reduce congestion of the communication channel and processing pressure on the processor during communication transmission.
[0025] In some embodiments, a camera or other image acquisition device is used to acquire raw video images of the foggy scene.
[0026] In this embodiment of the invention, each frame of the original video image undergoes format conversion, resolution scaling, warm-up frame control, frame-interval inference, and letterbox processing to obtain the target image frame. For example, the original video image in YUV420SP format is converted into the target image frame in RGB format to meet the input requirements of the subsequent recognition model. The image is scaled according to a set resolution to ensure that the target image frame is consistent with the input size of the recognition processing link. To avoid recognition anomalies caused by unstable buffering during the initial system startup, a warm-up frame mechanism is set up. This warm-up phase only performs video acquisition and encoding. The system adopts a frame-interval inference mechanism and does not perform... Instead of performing recognition processing on every frame of the original video image, a subset of video frames is selected for recognition according to a preset inference interval to reduce the computational burden on edge devices and improve real-time performance. Before being fed into the target detection model, the system performs letterbox processing on the image while maintaining its aspect ratio. This scales the original video image to the model's input size and fills blank areas at the edges. The scaling ratio and filling parameters are recorded. Letterbox processing is an image size standardization method that maintains the original image's aspect ratio, performs only proportional scaling and symmetrical filling, and its core principle is to avoid stretching and distortion. The recorded parameter information provides a basis for subsequent coordinate reconstruction. After a series of preprocessing steps, the target image frame is obtained.
[0027] In a specific embodiment of the present invention, for each fog scene, the target image frame is divided into multiple recognition areas according to a preset division rule (channel area or site), and the corresponding spray partition is controlled according to the recognition result in the recognition area.
[0028] In this embodiment of the invention, each recognition region is mapped to the video frame coordinate system, and boundary information is configured for each recognition region. For example, the boundary information of a rectangular region is (x1, y1, x2, y2).
[0029] S200 identifies target individuals in all recognition regions of the target image frame based on a preset detection model and outputs the recognition results; the recognition results include the target category, the confidence score and bounding box coordinates corresponding to each target category.
[0030] In this embodiment of the invention, the preset detection model is YOLOv5, YOLOv8, or other lightweight object detection models. In a preferred embodiment, the preset detection model is deployed using pre-trained weights combined with fine-tuning based on on-site samples to adapt to changes in lighting, background interference, and personnel distribution characteristics under different foggy scenes. Specifically: images or videos of the on-site foggy scene are collected as samples; the samples are labeled (target personnel bounding boxes and categories are labeled); the pre-trained model is fine-tuned using the labeled on-site samples, or data augmentation techniques are used to increase the number of samples and improve the model's generalization ability, thereby adapting to the recognition needs of different foggy scenes.
[0031] In this embodiment of the invention, the preset monitoring module outputs identification results including not only target category, target confidence level, and boundary coordinates, but also target center point coordinates, region number, or trajectory information as a control basis. For example, the identified target categories include personnel, vehicles, etc.
[0032] S300 processes the recognition results based on the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters.
[0033] In embodiments of the present invention, such as Figure 2 As shown, the recognition results are processed based on the parameter information recorded in the preprocessing process to obtain structured data, including: S310, based on the scaling ratio and fill parameters recorded in the Letterbox processing stage, reverse-calculates all bounding box coordinates to the original video image coordinates and crops them to the effective range.
[0034] Since the bounding box coordinates are initially set in the coordinate system of the preset detection model, rather than the coordinate system of the original video image, they cannot be directly used for spray control judgment. It is necessary to first restore the bounding box coordinates of the recognition results to the coordinate system of the original video image. After coordinate restoration, the bounding box coordinates are cropped to ensure that the position coordinates of each target category are within the valid range of the image, preventing coordinate overflow from affecting subsequent control judgments.
[0035] S320: Based on the target category and target confidence level in the identification results, select the valid target personnel and count the number of target personnel.
[0036] In this embodiment of the invention, valid target personnel are screened based on the target category and confidence threshold. Only the detection results that meet the target personnel category conditions and identification confidence requirements are retained. Then, the number of valid target personnel is counted to obtain the total number of people on site at the current time, i.e., the number of target personnel.
[0037] S330, determine the center point coordinates of the target person based on the bounding box coordinates of the target person.
[0038] In this embodiment of the invention, based on the bounding box coordinates of each valid target person, the system can further calculate the center point coordinates of the target person; as another implementation method, it can directly determine the spray zone to which each target person belongs.
[0039] S340 encapsulates the target personnel, the number of target personnel, the confidence level of the target personnel, the center point coordinates of the target personnel, and the timestamp of the target image frame into structured data.
[0040] In this embodiment of the invention, the target personnel, the number of target personnel, the target confidence level of the target personnel, the center point coordinates of the target personnel, and the timestamp of the target image frame obtained from the identification results are encapsulated into structured data in JSON or binary structures.
[0041] In this embodiment of the invention, structured data is transmitted to the control module via a bus, enabling the control module to perform spray control based on the structured data. Other implementation methods may also include transmitting structured data using shared memory, serial communication, network socket communication, MQTT, HTTP interface, or industrial bus.
[0042] This invention enables the decoupled deployment of the identification end and the control end through a message bus, allowing the identification module and the control module to run on the same device or on different devices, and to communicate through a unified data format.
[0043] The S400 generates corresponding control commands based on structured data and pre-set control modes; and controls the execution equipment of the corresponding spray zone according to the control commands.
[0044] In this embodiment of the invention, corresponding control commands are generated based on structured data and a pre-set control mode, including: (1) Determine the identification area to which the target person belongs based on the coordinates of the center point of the target person in the structured data; (2) Determine the regional status results of the same identification area based on the identification areas to which all target personnel belong; (3) Generate corresponding control commands based on the status results of all regions and the preset control mode.
[0045] For example, if the center point coordinates of a target person fall within a certain identification area, then all target persons in the same area are aggregated to generate an area status result (e.g., whether the area is occupied or unoccupied, the number of people in the area, etc.). Then, based on the area status result and the pre-set control mode, the control command for spraying is determined to control the opening or closing of the corresponding spray zone.
[0046] In this embodiment of the invention, the preset control mode includes whether to activate the follow-spray mode; If the follow spray mode is enabled, when the area status result of a certain identification area is "person", the corresponding spray zone of the identification area will be activated. If the follow spray mode is not enabled, when the area status result of any identified area is "occupied", control to enable all spray zones; As an alternative implementation, if the follow spray mode is not enabled, when the area status result of a certain identification area is "person", the spray partitions corresponding to all identification areas outside the identification area will be enabled.
[0047] For example, a fog generator controls multiple spray zones, each corresponding to a recognition area (such as area A, area B, and area C). When the coordinates of a target person are mapped to a corresponding recognition area (such as area A), if the control mode is set to "Follow Spray Mode," a control command is generated to activate only spray zone a corresponding to the target person's recognition area A. When the target person moves to recognition area B, a control command is generated to deactivate spray zone a and activate spray zone b corresponding to the target person's recognition area B; and so on. If the pre-set control mode is "Deactivate Follow Spray Mode," then when the target person arrives at recognition area A, spray zone a corresponding to recognition area A is deactivated, spray zone b corresponding to recognition area B and spray zone c corresponding to recognition area C are activated; when the target person arrives at recognition area B, spray zone b is deactivated, and spray zones a and c are activated.
[0048] In some preferred embodiments, if personnel are present in multiple adjacent identification areas at the same time, the control of adjacent spray areas can be activated in a coordinated manner.
[0049] In some preferred embodiments, in addition to "open when someone is present, close when no one is present", the control logic can also adopt triggering methods such as number threshold, dwell time, adjacent area linkage, or trajectory prediction, so as to make the control more precise, significantly reduce water consumption, power consumption and equipment ineffective operation time, and reduce system operating costs.
[0050] The examples above are for clearer illustration and do not constitute a limitation on the basis for generating control instructions of the present invention.
[0051] In this embodiment of the invention, the on / off switch of the corresponding spray zone is precisely controlled according to the identification area where the target person is located, based on the person, location, and area. This ensures timely spraying while avoiding continuous spraying in unoccupied areas and improving resource utilization efficiency.
[0052] In this embodiment of the invention, the actuating device includes a spray host, a solenoid valve, a zone valve, a relay pump control unit, or other switching actuators.
[0053] This allows the spray effect to dynamically change with people's activities, enhancing the interactive experience and intelligence of the fog scene.
[0054] Specifically, if no target person is detected in a certain identification area for several consecutive frames, or if no target person is detected again within a set delay period, the corresponding spray device will be turned off to avoid frequent spraying fluctuations due to occasional missed detections. When the system determines that there are no people in all identification areas, it can turn off all spray areas or switch to standby mode.
[0055] The automatic control method for fogging in this invention also pushes on-site video images to an upper-level platform, monitoring platform, or integrated management platform via the GB28181 protocol; it also supports real-time viewing of on-site video images via an App or Web interface, and can simultaneously display AI recognition results, spray area status, and equipment operating status.
[0056] This invention identifies the number and location of target personnel in a foggy scene in real time through video images, which has higher recognition accuracy and more complete control basis than traditional infrared sensing or timed control methods.
[0057] This invention also proposes a specific embodiment of an automatic control system for fog scenes, such as... Figure 3 As shown, the automatic control system of this fog scene includes a data acquisition module, a recognition module, a processing module, and a control module. The acquisition module is used to acquire the original video images of a preset area in the fog scene, and divide the corresponding target image frame obtained after preprocessing each original video image into multiple recognition areas; The recognition module is used to identify target individuals in all recognition regions of the target image frame based on a preset detection model, and outputs the recognition results. The recognition results include the target category, the confidence score for each target category, and the bounding box coordinates. The processing module is used to process the recognition results based on the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters; The control module is used to generate corresponding control commands based on structured data and pre-set control modes, and to control the execution equipment of the corresponding spray zone according to the control commands.
[0058] In this embodiment of the invention, the acquisition module uses a camera or other image acquisition device to acquire the original video image of a preset area of the fog scene. Then, each frame of the original video image is preprocessed to obtain the target image frame. The target image frame is then divided into multiple recognition areas according to a preset division rule.
[0059] In this embodiment of the invention, the preset monitoring model in the identification module includes YOLOv5, YOLOv8, or other lightweight target detection models. The preset monitoring module outputs identification results including not only the target category, target confidence score, and boundary coordinates, but also the target center point coordinates, region number, or trajectory information as a control basis. For example, the identified target categories include people, vehicles, etc.
[0060] In this embodiment of the invention, the processing module can be implemented using a PLC, a microcontroller, an industrial computer, an edge controller, or an embedded gateway.
[0061] In this embodiment of the invention, the actuating device includes a spray host, a solenoid valve, a zone valve, a relay pump control unit, or other switching actuators. The actuating device may employ solenoid valve control, relay control, variable frequency pump control, or a combination thereof.
[0062] The automatic control system for the fog scene of the present invention also includes a display module. The display module of the terminal platform is wirelessly connected to the acquisition module. The acquisition module transmits real-time video images of the fog scene to the terminal platform based on the GB28181 protocol and displays the video images through the display module.
[0063] In addition to the national standard GB28181 protocol, the video networking method of this invention can also adopt RTSP, RTMP, WebRTC, or other streaming media transmission methods. The automatic control system of this invention can achieve dynamic linkage control and remote visual viewing of spray zones according to changes in the location of the target personnel.
[0064] This invention also supports real-time viewing of on-site video footage via an app or web interface, and can simultaneously display AI recognition results, spray area status, and equipment operating status, thereby improving the efficiency of remote operation and maintenance.
[0065] This application also provides a computer-readable storage medium for storing the computer program used in the aforementioned terminal device. For example, the computer-readable storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0066] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that, as an alternative implementation, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.
[0067] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.
[0068] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, 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 smartphone, personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.
[0069] Finally, it should be noted that the above embodiments are only for illustration and not for limiting the technical solutions of the present invention. Any equivalent substitutions, modifications or partial substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. An automatic control method for a foggy scene, characterized in that, include: The original video images of a preset area of the fog scene are acquired, and the corresponding target image frames obtained after preprocessing each frame of the original video images are divided into multiple recognition areas; Based on a preset detection model, target personnel are identified in all recognition regions of the target image frame, and the recognition results are output. The recognition results include the target category, the confidence score for each target category, and the bounding box coordinates; The recognition results are processed based on the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters; the recognition results include target personnel; Based on the structured data and the pre-set control mode, corresponding control instructions are generated, and the execution devices of the corresponding spray zone are controlled according to the control instructions.
2. The automatic control method for fog scenes according to claim 1, characterized in that, The target image frame is obtained by performing format conversion, resolution scaling, and letterboxing on each frame of the original video image.
3. The automatic control method for fog scenes according to claim 1, characterized in that, The step of processing the recognition results based on the parameter information recorded in the preprocessing process to obtain structured data includes: Based on the scaling ratio and padding parameters recorded during the Letterbox processing stage, all bounding box coordinates are reverse-calculated to the original video image coordinates and cropped to the effective range. Based on the target category and target confidence level in the identification results, valid target personnel are selected and the number of target personnel is counted. The center point coordinates of the target person are determined based on the bounding box coordinates of the target person. The target personnel, the number of target personnel, the confidence level of the target personnel, and the coordinates of the center point of the target personnel are encapsulated into structured data.
4. The automatic control method for fog scenes according to claim 3, characterized in that, The step of generating corresponding control commands based on the structured data and the pre-set control mode includes: The identification area to which the target person belongs is determined based on the center point coordinates of the target person; Determine the regional status results for the same identification area based on the identification areas to which all target personnel belong; Based on the status results of all regions and the preset control mode, corresponding control commands are generated.
5. The automatic control method for fog scenes according to claim 4, characterized in that, The preset control modes include whether to enable the follow spray mode; If the follow spray mode is enabled, when the area status result of a certain identification area is "person", the corresponding spray zone of the identification area will be activated. If the follow spray mode is not enabled, when the area status result of any identified area is "occupied", control to activate all spray zones.
6. The automatic control method for fog scenes according to claim 4, characterized in that, If the number of target personnel in the area status result exceeds the preset number and the stay duration of the target personnel exceeds the preset duration, a control command to activate will be triggered.
7. An automatic control system for a fog scene, characterized in that, It includes a data acquisition module, an identification module, a processing module, and a control module; The acquisition module is used to acquire the original video images of a preset area of the fog scene, and divide the corresponding target image frame obtained after preprocessing each frame of the original video image into multiple recognition areas; The recognition module is used to identify target persons in all recognition regions of the target image frame based on a preset detection model, and output the recognition results; The recognition results include the target category, the confidence score for each target category, and the bounding box coordinates; The processing module is used to process the recognition results according to the parameter information recorded in the preprocessing process to obtain structured data; the parameter information includes scaling ratio and padding parameters; The control module is used to generate corresponding control commands based on the structured data and the pre-set control mode, and to control the execution equipment of the corresponding spray zone according to the control commands.
8. The automatic control system for fog scenes according to claim 7, characterized in that, It also includes a display module, which is wirelessly connected to the acquisition module of the terminal platform. The acquisition module transmits real-time video images of the fog scene to the terminal platform based on the GB28181 protocol and displays the video images through the display module.
9. The automatic control system for fog scenes according to claim 7, characterized in that, The actuators include a spray unit, a solenoid valve, a zone valve, a relay pump control unit, or other switching actuators.
10. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed on a processor, implements the automatic control method for fog scenes according to any one of claims 1-6.