A method and system for parallel processing of patrol tasks based on camera preset positions

Abstract

The present invention discloses a method for parallel processing of patrol tasks based on camera preset positions, including logging into a patrol task parallel processing system, a user adding a patrol plan; a user executing a patrol task or a system executing a task at a fixed time, and the system executing the task in parallel; a system interface obtaining patrol task information; judging whether the camera is a spherical camera, and performing algorithm recognition on the collected patrol data; a patrol report generation module performing statistical analysis on the patrol data, saving the patrol report results, and ending the patrol, which can improve the patrol efficiency and quality, can perform parallel patrols, and realize multi-scene patrol coverage; in addition, a patrol task parallel processing system based on camera preset positions is disclosed, including a camera, a communication module, a control module, a patrol report generation module, a storage module, a video monitoring module, and an intelligent patrol terminal, which improves the convenience and real-time performance of performing patrol tasks.

Description

Technical Field

[0001] The invention relates to the technical field of power plant inspection, and in particular to a method and system for parallel processing of inspection tasks based on camera preset positions. Background Art

[0002] With the rapid development of the national economy, the construction of power plants has also increased. As an important foundation for national energy security and the lifeline of the national economy, the problem of power grid safety management has become increasingly prominent. In order to ensure the safe operation of power plants, intelligent inspection management has been widely used to ensure the safe operation of power plants. At present, the intelligent inspection system of power plants includes the following: non-intelligent inspection management system: simply generates inspection tasks, issues inspection tasks, and performs manual inspections, which are purely manual inspections. Existing intelligent inspection management system: can basically realize inspections, but has a single function and lacks flexible configuration. The existing non-intelligent inspection system requires a system to generate inspection tasks and notify personnel to conduct inspections. It cannot achieve automatic inspections of the system, and has poor security and supervision.

[0003] The disclosure discloses an intelligent inspection system and an intelligent inspection method with publication number CN104157021B. The system comprises an intelligent inspection terminal, an inspection server and an inspection management center. The intelligent inspection terminal and the inspection server are connected via a wireless network, and the inspection server and the inspection management center are connected via TCP / IP. The intelligent inspection terminal is a GPS inspection terminal, comprising a shell, a display arranged on the front of the shell, a positioning module inside the shell, an audio / video module, a GIS map module, an emergency plan management module, a power module, an input module and a transmission module. The positioning module, the GIS map module, the emergency plan management module and the audio / video module are all connected to the display, the transmission module and the power module, and the display, the transmission module and the power module are connected to each other. However, the inspection object of the intelligent inspection system is single, and it is impossible to achieve coverage of multi-scenario inspections. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method and system for parallel processing of patrol tasks based on camera preset positions, which can achieve multi-scene patrol coverage and improve patrol efficiency and quality.

[0005] The objective of the present invention is achieved through the following technical solutions:

[0006] A method for parallel processing of patrol tasks based on camera preset positions, comprising:

[0007] Step 1: Log in to the inspection task parallel processing system and add an inspection plan;

[0008] Step 2: The user performs the inspection task or the system performs the task regularly, and the system performs the task in parallel;

[0009] Step 3: The system interface obtains inspection task information;

[0010] Step 4: determine whether the camera is a spherical camera. If so, further determine whether the camera is occupied; if not, call the camera to take a photo;

[0011] Step 5: Set multiple cameras to execute AI recognition algorithms at different preset positions simultaneously for parallel inspection;

[0012] Step 6: The inspection report generation module performs statistical analysis on the inspection data, saves the inspection report results, and the inspection ends.

[0013] Furthermore, the step 4 also includes waiting for the camera lock to be released if the camera is occupied; if not occupied, sending a camera call instruction to make the camera move to the corresponding preset position.

[0014] Furthermore, the method also includes regional classification management of cameras, real-time monitoring and playback of historical videos, and querying related video information based on cameras, distribution areas, and time conditions. The distribution areas correspond to cameras, and when there are faulty screens, they are automatically popped up for display quickly and timely to meet the playback of single-screen, 4-screen, 9-screen multi-channel video recordings, and to grasp the on-site situation from multiple angles.

[0015] Furthermore, the inspection tasks include personnel detection, safety helmet recognition, work clothes recognition, license plate recognition, and meter recognition, involving inspection personnel, inspection routes, inspection objects, inspection equipment, inspection start time, and inspection end time.

[0016] Furthermore, the method also includes customizing patrol functions for different patrol groups and different cameras. During patrol, the camera's screen is displayed as any single screen or multiple screens. The patrol function includes automatic timed patrol and manual patrol. The patrol interval can be set. The camera screen freezes to display the current video, and multiple camera signals are transmitted, switched and displayed, and stored as a backup in real time.

[0017] A patrol task parallel processing system based on camera preset positions includes a camera for acquiring identification information; a communication module for transmitting video data, personnel data, and vehicle data acquired by the camera; a control module for controlling the camera's all-round angles; an inspection report generation module for performing statistical analysis on inspection data and generating corresponding inspection reports; a storage module for storing videos and inspection data acquired by the camera; a video monitoring module for collecting and collating inspection data acquired by the camera; and an intelligent inspection terminal for viewing on-site video images and inspection data. The communication module transmits inspection reports to the intelligent inspection terminal.

[0018] Furthermore, the intelligent inspection terminal includes a video preview page, which can retrieve and preview the monitoring video in real time, and the video preview page can capture the on-site video screen in real time.

[0019] Furthermore, the communication module is optical fiber, 4G, LoRA, ZIGBEE or WIFI.

[0020] Furthermore, the system also includes a voice module for voice reminder of inspection report results.

[0021] Furthermore, the system also includes an inspection management center, which locates inspection personnel in real time and records and compiles statistics on whether the inspection personnel have missed inspections, inspected late, deviated from the inspection route, or have incorrect inspection sequences; the inspection management center adds and modifies emergency plans based on the statistics of detection data within the power plant. The emergency plans are in the form of text, pictures, audio / video, and are available for download by smart inspection terminals.

[0022] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0023] (1) The present invention can avoid the uncertainty of manual inspection, improve inspection efficiency and quality, and support real-time and scheduled mixed inspections. It can conduct parallel inspections and inspect multiple tasks at the same time, thus achieving multi-scenario inspection coverage.

[0024] (2) It can manage inspection tasks, configure corresponding inspection objects and corresponding inspection cameras for inspection tasks, and flexibly combine them to conduct inspection and monitoring of the on-site environment;

[0025] (3) All inspection tasks can be planned and triggered on time by configuring the time, thus avoiding manual inspection triggering and improving the convenience and real-time performance of inspection tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of a method for parallel processing of patrol tasks based on camera preset positions;

[0027] Figure 2 This is a module diagram of a patrol task parallel processing system based on camera preset positions. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] like Figure 1 As shown, a method for parallel processing of patrol tasks based on camera preset positions includes:

[0030] Step 1: Log in to the inspection task parallel processing system, and the user adds an inspection plan, which can be customized;

[0031] Step 2: The user executes the inspection task or the system executes the task regularly, and the system executes the tasks in parallel, and the number of tasks is greater than or equal to 1;

[0032] Step 3: The system interface obtains inspection task information;

[0033] Step 4: determine whether the camera is a spherical camera. If so, further determine whether the camera is occupied; if not, call the camera to take a photo;

[0034] Step 5: Set multiple cameras to execute AI recognition algorithms at different preset positions simultaneously for parallel inspection;

[0035] Step 6: The inspection report generation module performs statistical analysis on the inspection data, saves the inspection report results, and the inspection ends.

[0036] Parallel processing is a computing method in which two or more processes are executed simultaneously in a computer system. Parallel processing allows different aspects of the same program to be worked on at the same time. The main purpose of parallel processing is to save time in solving large and complex problems. In order to use parallel processing, the program must first be parallelized, that is, the work must be divided into different processes.

[0037] Step 4 also includes waiting for the camera lock to be released if the camera is occupied; if not occupied, sending a camera call instruction to move the camera to the corresponding preset position.

[0038] in,

[0039] The method also includes regional classification management of cameras, real-time monitoring and playback of historical videos, and querying related video information according to the cameras, distribution areas, and time conditions. The distribution areas correspond to the cameras. When there is a faulty screen, it is quickly and timely popped up and displayed automatically to meet the playback of single-screen, 4-screen, and 9-screen multi-channel video recordings, so that the on-site situation can be grasped from multiple angles. It can be seamlessly connected to the LCD large screen, and the video screen can be retrieved and displayed on the large screen system conveniently, quickly, intuitively, and smoothly.

[0040] Inspection tasks include personnel detection (face recognition attendance), helmet recognition (whether the helmet is worn correctly), work clothes recognition (whether the work clothes are worn correctly), license plate recognition (vehicle model recognition and unloading volume statistics at the crushing station unloading port, license plate recognition), meter recognition (various pointer meter recognition), involving inspection personnel, inspection routes, inspection objects, inspection equipment, inspection start time, and inspection end time.

[0041] Among them, the personnel and vehicle management in the inspection task is to effectively manage the personnel and vehicles in each monitoring area. Personnel and vehicle management includes personnel and vehicle submodules. The personnel submodule displays the management of face capture images when people enter and exit each monitoring area. The page content includes information such as monitoring equipment, monitoring area, face image, and image capture recording time. The entry and exit of personnel can be queried based on monitoring equipment, monitoring area, recording time and other conditions to improve the safety of personnel in the power plant and the supervision of the power plant.

[0042] Vehicle submodule display and external vehicle in and out information management. The page content includes the license plate, channel name, in and out type, vehicle type (factory and external), in and out time and other information of the factory and external vehicles. The vehicle in and out situation can be queried based on the license plate, channel name, in and out type, vehicle type and other conditions.

[0043] In step 5, the inspection report generation module performs statistical analysis on the inspection data, which can be managed by AI algorithm. AI algorithm management is to effectively manage various artificial intelligence (AI) algorithm recognition information, including independent sub-modules of each algorithm, namely personnel intrusion recognition, vehicle information recognition, helmet recognition, tooling recognition, meter recognition, etc. The system function interface style of each sub-module is consistent. Each sub-module displays the management of the equipment captured picture information when each monitoring device intelligently recognizes an abnormal situation. The sub-module consists of three parts: data analysis, dynamic curve, and latest dynamics. Data analysis displays the statistical data of daily, monthly, and annual captured pictures; the dynamic curve displays the statistical data curve of the captured pictures of each monitoring device in the past 7 days / 30 days / 90 days; the latest dynamics displays all captured pictures today, supports batch management of the status of all pictures (uncertain, confirmed, false alarm), and can query the captured picture information by date, monitoring equipment and other conditions.

[0044] The method also includes customizing different patrol groups and different cameras to form patrol functions. During patrol, the camera screen is displayed as any single screen or multiple screens. The patrol function includes automatic timed patrol and manual patrol, setting the patrol interval time, the camera screen freezes to display the current video, and real-time transmission, switching display, and backup storage of multi-channel camera signals.

[0045] like Figure 2As shown, a patrol task parallel processing system based on camera preset positions includes a camera for obtaining identification information; a communication module for transmitting video data, personnel data, and vehicle data obtained by the camera; a control module for controlling the camera's all-round angles; an inspection report generation module for performing statistical analysis on the inspection data and generating corresponding inspection reports; a storage module for storing videos and inspection data collected by the camera; a video monitoring module for collecting and collating the inspection data collected by the camera; and an intelligent inspection terminal for viewing on-site video images and inspection data. The communication module transmits the inspection report to the intelligent inspection terminal, which can be a PC, a large LCD display, a mobile phone, and the like.

[0046] The intelligent inspection terminal includes a video preview page, which can be used to preview surveillance videos in real time. The video preview page can capture the live video in real time and can replay the video of the selected time zone according to the selected time period. The playback video has the functions of fast forward, slow forward and pause. The preview video page has the function of real-time capture and adjustable sound volume; the video recording page can replay the video of the selected time zone according to the selected time period. The playback video has the functions of drive-in, capture, pause, etc.

[0047] Among them, the communication modules are optical fiber, 4G, LoRA, ZIGBEE or WIFI. When 5G networks become popular in the future, the system can expand its support for 5G networks. This system has highly compatible access capabilities.

[0048] The system also includes a voice module, which can remind users of the inspection report results. When the video surveillance module detects that someone has mistakenly entered the power plant or the vehicle information is inconsistent, the voice module can issue an alarm to remind the user to take the next step in handling the emergency. In addition, the system can also send a command to start a drone inspection.

[0049] The system also includes an inspection management center, which locates inspection personnel in real time and records and compiles statistics on whether the inspection personnel have missed inspections, inspected late, deviated from the inspection route, or inspected in the wrong order. The inspection management center adds and modifies emergency plans based on the statistics of inspection data within the power plant. The emergency plans are in the form of text, pictures, audio / video, and are available for download by smart inspection terminals.

[0050] The system supports intelligent inspection operations for registered monitoring devices, automatically obtains inspection results on demand, and exports inspection reports for retrieval and review. At the same time, the system supports specifying different inspection methods, including manual, automatic, and periodic. It supports highly customized and flexible expansion of inspection strategies, up to 256 inspection strategies, and provides editing, definition, and combination of camera preset levels to meet customized inspection requirements.

[0051] Embodiment 1

[0052] The user logs into the system, adds inspection plans, and customizes inspection routes. By triggering inspection tasks, the system executes inspection tasks in parallel and sends instructions to call related cameras for inspection. Cameras are set at multiple preset positions, and each camera has different inspection tasks. Through multiple cameras, pictures are taken for inspection and AI algorithm intelligent recognition is performed at the same time. Among them, the first inspection task is: vehicle identification. The license plate recognition algorithm is built into the front-end device. The license plate information is identified in the access monitoring area, and the access of the factory and external vehicles is mastered, and the access of vehicles is effectively managed. The vehicle identification system is installed at the unloading port of the crushing station to realize the model identification and unloading quantity statistics of mining trucks of various tonnages. The second inspection task is: face recognition, taking pictures of the staff entering and exiting the monitoring area to identify the face information; the third inspection task is: tooling identification; the fourth inspection task is: meter identification. The number of inspection tasks can be processed in parallel according to the actual situation. The inspection report generation module performs statistical analysis on the inspection data, and the system saves the recognition results to complete the inspection.

[0053] Embodiment 2

[0054] The patrol task parallel processing system controls the camera's all-round angles through the camera's control module to simultaneously shoot and obtain identification information, and conduct parallel patrol inspections. The video monitoring module collects and organizes patrol data. The communication module then transmits on-site video data, personnel data, vehicle data, etc. to the patrol report generation module, performs statistical analysis on the patrol data, and generates corresponding patrol reports. The storage module stores the data, and the patrol report is then transmitted to the intelligent patrol terminal through the communication module for users to view, thereby improving patrol efficiency and quality.

[0055] It should be noted that the above-mentioned specific implementation methods are only preferred embodiments of the present invention and the technical principles used. Within the technical scope disclosed by the present invention, any changes, modifications, substitutions, combinations and simplifications made by any technician familiar with the technical field without departing from the spirit and principles of the present invention should be equivalent replacement methods and should be covered by the protection scope of the present invention.

Claims

1. A method for parallel processing of patrol tasks based on camera preset positions, characterized in that: include: Step 1: Log in to the inspection task parallel processing system and add an inspection plan; Step 2: The user performs the inspection task or the system performs the task regularly, and the system performs the task in parallel; Step 3: The system interface obtains inspection task information; Step 4: determine whether the camera is a spherical camera. If so, further determine whether the camera is occupied; if not, call the camera to take a photo; Step 5: Set multiple cameras to execute AI recognition algorithms at different preset positions simultaneously for parallel inspection; Step 6: The inspection report generation module performs statistical analysis on the inspection data, saves the inspection report results, and the inspection is completed. The step 4 also includes waiting for the camera lock to be released if the camera is occupied; If it is not occupied, send a camera call command to move the camera to the corresponding preset position. The method also includes customizing different patrol groups and different cameras to form patrol functions. During patrol, the camera screen is displayed as any single screen or multiple screens. The patrol function includes automatic timed patrol and manual patrol. The patrol interval is set. The camera screen freezes to display the current video, and real-time transmission, switching display, and backup storage of multi-channel camera signals are performed. The method also includes regional classification management of cameras, real-time monitoring and playback of historical videos, and querying related video information based on cameras, distribution areas, and time conditions. The distribution areas correspond to cameras, and when there are faulty screens, they are quickly and automatically popped up for display, meeting the playback of single-screen, 4-screen, 9-screen multi-channel video recordings, and grasping the on-site situation from multiple angles.

2. The method for parallel processing of patrol tasks based on camera preset positions according to claim 1 is characterized in that: The inspection tasks include personnel detection, safety helmet recognition, work clothes recognition, license plate recognition, and meter recognition, involving inspection personnel, inspection routes, inspection objects, inspection equipment, inspection start time, and inspection end time.

3. A patrol task parallel processing system based on camera preset positions according to any one of claims 1 to 2, characterized in that: It includes a camera for acquiring identification information; a communication module for transmitting video data, personnel data, and vehicle data acquired by the camera; and a control module for controlling the camera's full range of angles; Inspection report generation module, used to perform statistical analysis on inspection data and generate corresponding inspection reports; Storage module, used to store video and inspection data collected by the camera; Video monitoring module, used to collect and organize inspection data collected by cameras; The intelligent inspection terminal is used to view on-site video images and inspection data, and the communication module transmits the inspection report to the intelligent inspection terminal.

4. The patrol task parallel processing system based on camera preset positions according to claim 3 is characterized in that: The intelligent inspection terminal includes a video preview page, which can retrieve and preview monitoring videos in real time, and the video preview page can capture live video images in real time.

5. The patrol task parallel processing system based on camera preset positions according to claim 4 is characterized in that: The communication module is optical fiber, 4G, LoRA, ZIGBEE or WIFI.

6. The patrol task parallel processing system based on camera preset positions according to claim 5 is characterized in that: The system also includes a voice module for voice reminder of inspection report results.

7. The patrol task parallel processing system based on camera preset positions according to claim 6 is characterized in that: The system also includes an inspection management center, which locates the inspection personnel in real time and records and counts whether the inspection personnel have missed inspections, inspected late, deviated from the inspection route, or have made inspections in the wrong order; The inspection management center adds and modifies emergency plans based on the statistics of detection data within the power plant. The emergency plans are in the form of text, pictures, audio / video, and are available for download by smart inspection terminals.

Images