Factory intelligent monitoring system
By deploying intelligent monitoring systems in railway plants and road crossings, and using AI technology to monitor and alarm in real time, the problem of difficulty in effectively monitoring safety in traditional manual patrols has been solved, and production safety guarantee and accident prevention capabilities have been improved.
Patent Information
- Application Number
- CN202421958031.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2025-06-17
- Estimated Expiration
- 2034-08-13
AI Technical Summary
In the railway industry production, traditional personnel patrol methods are difficult to effectively monitor the safety of the factory area and road crossing, resulting in the occurrence of potential risk factors and safety accidents.
An intelligent monitoring system in the factory was designed, including a server, a video monitoring module in the factory, an intelligent monitoring module in the roadway, and a perimeter intrusion alarm module. Video images are collected and transmitted through cameras, switches and distribution boxes, and real-time monitoring and alarm are used to use AI image analysis programs.
All-weather monitoring of railway plant areas, crossings and perimeters has been achieved, production safety guarantees have been improved, human resource waste has been reduced, and safety accidents have been effectively prevented.
Smart Images
Figure CN222996606U_ABST
Abstract
Description
Technical Field
[0001] The utility model belongs to the field of railway factory area crossing monitoring, and particularly relates to a monitoring system. Background Art
[0002] In railway industrial production, factories and production facilities usually involve potential risk factors such as large machinery, high temperature, high pressure, and chemical substances. Accidents should be avoided as much as possible. Personnel management is also an important aspect. How to ensure that employees comply with work procedures and safety standards is currently achieved through manual persuasion without a unified standard. Perimeter security is very important. Traditional personnel patrols are time-consuming and laborious. The factory area is connected to the railway, inevitably forming a complex traffic facility such as a crossing. Usually, the crossing is manned manually, which wastes a large amount of human resources. In such a complex situation, accidents may occur. Content of the Utility Model
[0003] To solve the above technical problems, the utility model proposes an intelligent monitoring system for the factory area.
[0004] The technical solution adopted by the utility model is as follows: The intelligent monitoring system for the factory area includes: a server, a factory area video monitoring module, a crossing intelligent monitoring module, and a perimeter intrusion alarm module; the factory area video monitoring module, the crossing intelligent monitoring module, and the perimeter intrusion alarm module are connected to the server by wired or wireless means;
[0005] The factory area video monitoring module includes: a first camera, a first switch, and a first distribution box distributed in the factory area; the factory area video images are collected by the first camera distributed in the factory area, and the collected factory area video images are sent to the server through the first switch; the first distribution box is electrically connected to the first camera and the first switch;
[0006] The crossing intelligent monitoring module includes: a second camera, a second switch, a crossing barrier machine, a sound and light alarm, a fill light, and a second distribution box distributed at the crossing; the crossing video images are collected by the second camera distributed at the crossing, and the collected crossing video images are sent to the server through the second switch; the sound and light alarm includes: a sound alarm circuit and a buzzer; the second distribution box is electrically connected to the second camera, the second switch, the crossing barrier machine, the sound and light alarm, and the fill light;
[0007] The second camera, the second switch, the sound and light alarm, the fill light, and the second distribution box are installed on a pole. The number of fill lights matches the number of second cameras, and the irradiation angles of the installed fill lights are the same as the shooting angles of the corresponding second cameras.
[0008] One vertical pole is set on each side of each crossing; two second cameras are installed on each vertical pole. One of the two second cameras is responsible for monitoring the crossing area, and the other camera is responsible for monitoring the direction of the incoming and outgoing trains.
[0009] A signal lamp is also installed on each vertical pole, and the signal lamp is electrically connected to the second distribution box installed on the vertical pole.
[0010] The perimeter intrusion alarm module includes: the third cameras, the third switches, and the third distribution boxes distributed on the perimeter of the factory area; the video images of the perimeter of the factory area are collected by the third cameras distributed on the perimeter of the factory area, and the collected video images of the perimeter of the factory area are sent to the server through the third switches; the third distribution boxes are electrically connected to the third cameras and the third switches.
[0011] The beneficial effects of the present utility model: The intelligent monitoring system for the factory area of the present utility model includes: a server, a factory area video monitoring module, a crossing intelligent monitoring module, and a perimeter intrusion alarm module; the factory area video monitoring module, the crossing intelligent monitoring module, and the perimeter intrusion alarm module are connected to the server by wired or wireless means; the factory area video monitoring module, the crossing intelligent monitoring module, and the perimeter intrusion alarm module;
[0012] The factory area video monitoring module sends the collected video images of the factory area to the server through the cameras, switches, and distribution boxes distributed in the factory area;
[0013] The crossing intelligent monitoring module sends the collected video images of the crossing to the server through the cameras, switches, and distribution boxes distributed at the crossing; the server processes train reception and passing based on the video images of the crossing;
[0014] The perimeter intrusion alarm module sends the collected video images of the perimeter of the factory area to the server through the cameras, switches, and distribution boxes distributed on the perimeter of the factory area;
[0015] In summary, the monitoring system of the present utility model can effectively monitor the railway factory area and provide guarantee for the production safety of the railway factory area. Description of the Drawings
[0016] Figure 1 It is the intelligent monitoring system for the factory area of the present utility model.
[0017] Figure 2 It is the flowchart of train reception and passing at the crossing. Detailed Embodiment
[0018] To facilitate the understanding of the technical content of the present utility model by those skilled in the art, the content of the present utility model will be further explained below with reference to the drawings.
[0019] The intelligent video monitoring system for the factory area of the present utility model is applied to railway industrial production. The present utility model divides the railway production area into: a production operation area, a crossing area, and a factory area perimeter area. By installing cameras in these three areas, video data of each area is collected, and then advanced computer vision technology combined with artificial intelligence technology is used to realize the monitoring and management of the factory area, perimeter, and crossing.
[0020] The intelligent monitoring system for the factory area of the present utility model consists of four main parts: the front end, transmission, control, and display and recording. The front-end high-definition cameras are distributed on each floor of the factory area, perimeter, crossing, maintenance depot, and workshop according to the technical prevention requirements and the layout principles for different areas, and are transmitted to the monitoring center through the access layer switch of the security system private network; the monitoring center is configured with a data server, an image server, a video server, a video storage server, and a remote control device for the crossing, and also includes devices such as switches, monitors, and network cabinets.
[0021] Specifically:
[0022] The intelligent monitoring system platform for the factory area is deployed in the data server; an intelligent video analysis program is deployed in the image server. Based on the existing known AI image analysis program, real-time processing and analysis of various violation behaviors are carried out, such as functions of personnel detection, vehicle detection, and violation behavior analysis. Personnel detection, vehicle detection, and violation behavior analysis are carried out through the AI image analysis program, which is an existing known technology, and the present utility model will not elaborate on this here.
[0023] The integrated security management platform is deployed in the video server, which is responsible for summarizing and managing all monitoring devices and video stream encoding and decoding conversion in the entire factory area, perimeter, and crossing; the converted video data is used for analysis and detection by the image server and stored by the storage server. The video server is connected to the data server, video storage server, and image server through Ethernet cables.
[0024] The video storage server is responsible for storing the video data of all monitoring cameras in the entire factory area and the alarm data. Video storage is an existing known technology, and the present utility model will not elaborate on this here.
[0025] The remote control device for level crossings remotely controls the lifting rods of the barrier machines and the switches of the alarms at all level crossings through the network. The remote control device for level crossings consists of three parts: a network control module, which is installed in the distribution box on the level crossing pole. One end is connected to the barrier machine and the audible and visual alarm to enable remote control of the barrier machine through the network, and the other end is connected to the switch in the distribution box; the remote control device for the level crossing barrier machine, which is installed in the network cabinet in the central computer room and connected to the core switch to achieve remote network connection with all level crossing devices; the remote control handle device for the level crossing barrier machine, which is installed in the monitoring duty room and connected to the switch, forming an internal local area network with the network control module at the level crossing and the remote control device for the level crossing barrier machine in the computer room to achieve remote network connection and control functions for all level crossing devices.
[0026] The intelligent monitoring subsystem for level crossings conducts video surveillance on the level crossing area, and intelligently identifies pedestrians, vehicles, trains and other obstacles in the level crossing area through existing known AI image recognition technology. It real-time monitors the occupancy status of the level crossing, the status of trains entering the level crossing area and the status of trains leaving the level crossing area, and automatically controls the opening and closing of the barriers. For the occupancy of the level crossing after a train approaches, it gives an audible and visual alarm to prompt getting off the track or manual intervention and records it to ensure the safe passage of trains and prevent the occurrence of level crossing traffic accidents.
[0027] The perimeter intrusion alarm subsystem conducts video surveillance on the perimeter of the station yard, and intelligently identifies people, vehicles, etc. that break into the perimeter through existing known AI image recognition technology. It real-time monitors the perimeter intrusion status and automatically controls the pan-tilt for cruise surveillance. It automatically tracks, manually intervenes and records alarms for people, vehicles, environmental anomalies, etc. that invade the perimeter to ensure the safety of the station yard environment.
[0028] The monitoring subsystem in the maintenance depot conducts real-time monitoring of the maintenance situation in the maintenance depot, detects whether the relevant operations, clothing, and safety protection of the operating personnel comply with the regulations, and manually intervenes and gives an alarm for non-compliance to ensure the production safety of the relevant personnel in the maintenance depot.
[0029] The intelligent video monitoring system for the factory area provides operations such as the electronic map of the station yard, real-time preview of video monitoring, video playback and free download, intelligent monitoring of level crossings, perimeter intrusion alarm, monitoring in the maintenance depot and operation evaluation, search and query of alarm records, and system management (including: personnel management, permission management, equipment management, etc.), for the administrator to view the on-site situation of the station yard perimeter at any time to ensure the safety of the station yard environment.
[0030] The intelligent video monitoring system for the factory area includes: a video monitoring subsystem, an intelligent monitoring subsystem for level crossings, a perimeter intrusion alarm subsystem, and a monitoring and operation evaluation subsystem in the maintenance depot. The overall architecture diagram is as Figure 1 shown.
[0031] 1. Video monitoring subsystem
[0032] Video surveillance facilities are set up according to technical security requirements and for different areas, and are generally distributed on each floor of the factory area and the workshops, used for non-blind-spot monitoring of personnel and vehicles. It consists of cameras in the workshops, cameras in the factory area, disk arrays, and switches.
[0033] The video surveillance subsystem mainly completes the high-definition video stream acquisition and local video storage on each floor of the factory area and the workshops. The system uses digital high-definition cameras and disk arrays. The front-end devices of the cameras in the factory area and the workshops collect image signals and transmit them to the disk arrays for storage through switches. 2 million spherical high-definition infrared cameras are used in the factory area, 2 million hemispherical high-definition infrared cameras are used in rooms and offices in the workshops, and 2 million bullet-type high-definition infrared cameras are used in fixed areas such as corridors and hallways.
[0034] The video surveillance subsystem can perform real-time on-demand viewing of images according to specified devices and channels, support the display, zoom, capture, and recording of on-demand images, and support multiple users to simultaneously view the same image resource.
[0035] The video surveillance subsystem can remotely control various actions of the front-end devices through manual or automatic operations; it can set control priorities and have corresponding measures to ensure priority response to high-level user requests.
[0036] The video surveillance subsystem can retrieve historical image data and play back and download it according to elements such as specified devices, channels, time, and alarm information; playback supports normal playback, fast playback, picture pause, and image capture.
[0037] The master control center of the video surveillance subsystem is deployed using IP-SAN (disk array) storage, and the stored video information is not less than 30 days.
[0038] The video surveillance subsystem logs include two types: operation logs and running logs. The running logs can record the startup, self-check, anomalies, faults, recovery, shutdown, etc. status and occurrence times of the devices in the system; the operation logs can record the entry and exit times of operators and their main operation situations.
[0039] The video surveillance subsystem supports the function of detecting abnormal states of cameras. It can detect various abnormal states of cameras, including video line disconnection, camera damage, and camera movement.
[0040] 2. Roadway monitoring subsystem
[0041] The intelligent monitoring subsystem for level crossings supports round-the-clock and non-blind-spot monitoring of the level crossing area. The videos are transmitted through a switch to a disk array server for storage, and real-time preview and playback are also available. 2 million spherical high-definition cameras with intelligent light-controlled LED compensation light sources are used at level crossings for AI image recognition at night. The images are uploaded to an AI image analysis server for analysis.
[0042] The intelligent monitoring subsystem for level crossings supports functions such as oncoming vehicle warning for level crossings, audible and visual alarms for intrusion / occupation of the road, and automatic barrier control. A vehicle-receiving camera is installed 150 meters away from the level crossing. When the vehicle-receiving camera detects an oncoming vehicle, the audible and visual alarm at the level crossing gives an oncoming vehicle warning, and the AI image recognition program starts. It continuously monitors whether there are any stranded people or vehicles inside the level crossing. If there are people or vehicles stranded inside the level crossing, the barrier will not close and an alarm will be triggered. The alarm information will be uploaded to the alarm function module of the platform, and the voice message "Crossing the boundary, please get off the road immediately" will be played. When the people have left, the barrier will be closed to isolate the track from the level crossing. After the vehicle has passed, the barrier will be opened for free passage.
[0043] If the distance between adjacent level crossings is too close, a linkage method is adopted for the vehicle-receiving and vehicle-passing functions of the level crossings, as specifically shown in Figure 2 the following.
[0044] 3. Perimeter intrusion alarm subsystem
[0045] The perimeter intrusion alarm subsystem supports round-the-clock and non-blind-spot monitoring of the factory perimeter area. The cameras at the factory perimeter monitor the factory perimeter environment round-the-clock and without blind spots. The videos are transmitted through a switch to a disk array for storage, and real-time preview and playback are also available. 2 million spherical high-definition cameras are used in the perimeter area.
[0046] The perimeter intrusion alarm subsystem supports round-the-clock intelligent intrusion alarm functions for the factory perimeter, including detection and alarm for crossing the virtual warning wall; detection and alarm for regional intrusion; detection and warning for loitering; detection and warning for people gathering; detection and warning for the presence or loss of items in key areas.
[0047] Crossing detection: In the armed area, if a person crosses the boundary, AI image recognition will trigger an alarm. The alarm information will be uploaded to the alarm function module of the platform.
[0048] Regional intrusion detection: In the armed area, when a person enters the area, AI image recognition will trigger an alarm. The alarm information will be uploaded to the alarm function module of the platform.
[0049] Loitering detection: In the armed area, when a person enters the area multiple times and repeatedly, AI image recognition will trigger an alarm. The alarm information will be uploaded to the alarm module of the platform.
[0050] Personnel gathering: In the defense area, if a certain number of people reach and do not leave the area, staying for a long time, AI image recognition will give an alarm. The alarm will be uploaded to the platform alarm module
[0051] Items left or lost in key areas: In the defense area, if an item is left or lost, AI image recognition will give an alarm. The alarm will be uploaded to the platform alarm module.
[0052] 4. Monitoring subsystem in the maintenance depot (i.e., plant video monitoring module)
[0053] The monitoring subsystem in the maintenance depot supports round-the-clock video surveillance without blind spots in the maintenance depot. The cameras in the maintenance depot will monitor the environment and personnel in the maintenance depot round-the-clock without blind spots, and transmit the video to the disk array for storage through a switch. It can also be used for real-time preview and playback. 2 million spherical high-definition cameras are used in the maintenance depot, and the images are uploaded to the AI image analysis server for analysis.
[0054] The monitoring subsystem in the maintenance depot supports functions such as detecting violations of crossing the line, not wearing a safety helmet, not wearing work clothes, not wearing a mask, etc. in the maintenance depot.
[0055] Violation of crossing the line: Before the maintenance train enters the maintenance depot and comes to a complete stop, no personnel are allowed to step into the boundary. If someone enters the boundary, AI image recognition will give an alarm, upload it to the platform alarm function module, and remind the personnel who cross the line through an audible and visual alarm.
[0056] Not wearing a safety helmet: In the maintenance depot, it is stipulated that personnel in the depot must wear a safety helmet. If not wearing a safety helmet, AI image recognition will give an alarm and upload it to the platform alarm function module.
[0057] Not wearing work clothes: In the maintenance depot, it is stipulated that personnel in the depot must wear work clothes. If not wearing work clothes, AI image recognition will give an alarm and upload it to the platform alarm function module.
[0058] Mask recognition: In the maintenance depot, since harmful dust such as iron filings will be generated during the operation process, operators must wear masks. If not wearing a mask, AI image recognition will give an alarm and upload it to the platform alarm function module.
[0059] When an alarm event occurs, the system will display the alarm information on the web platform within 2 seconds for alarm prompt, and distribute the alarm information to the corresponding systems and devices for processing according to the alarm handling strategy. The latest alarm information can be viewed in the alarm list, and alarm playback, alarm event query, alarm electronic map prompt, alarm log management, and alarm data statistical analysis can be performed.
[0060] When the camera connection is interrupted, an alarm message for the camera connection interruption will be triggered, and the relevant alarm message will pop up on the web page. The alarm list will display the time, camera location information, and alarm type, facilitating maintenance by relevant personnel.
[0061] After the approaching vehicle warning, the crossing is about to close, and the AI image recognition program is started. If people and vehicles are detected inside the crossing, an alarm message for personnel intrusion will be triggered, and the relevant alarm message will pop up on the web page. The alarm list will display the time, crossing number, camera location, and alarm type, facilitating viewing by relevant personnel.
[0062] After the perimeter intrusion alarm camera is successfully deployed, the AI image recognition program is started. If an alarm of a relevant type is detected, a perimeter intrusion alarm type (crossing detection, area intrusion detection, loitering detection, people gathering, item left or lost in a key area) will be triggered, and the relevant alarm message will pop up on the web page. The alarm list will display the time, camera location, and alarm type.
[0063] After the cameras in the maintenance depot are successfully deployed, the AI image recognition program is started. If an alarm of a relevant type is detected, a monitoring and operation evaluation alarm type in the maintenance depot (illegal crossing of the line, not wearing a safety helmet, not wearing work clothes, not wearing a mask) will be triggered, and the relevant alarm message will pop up on the web page. The alarm list will display the time, camera location, and alarm type.
[0064] Those of ordinary skill in the art will realize that the embodiments described herein are to assist the reader in understanding the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. For those skilled in the art, various changes and modifications can be made to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.
Claims
1. The factory intelligent monitoring system is characterized by: include: Server, factory video surveillance module, road crossing intelligent monitoring module, perimeter intrusion alarm module; The factory area video surveillance module, the intelligent crossing surveillance module, and the perimeter intrusion alarm module are connected to the server via wired or wireless means; The factory video monitoring module includes: a first camera distributed in the factory, a first switch, and a first distribution box; the factory video image is collected by the first camera distributed in the factory, and the collected factory video image is sent to the server through the first switch; the first distribution box is electrically connected to the first camera and the first switch; The intelligent monitoring module for the road crossing includes: a second camera distributed at the road crossing, a second switch, a road crossing barrier, an audible and visual alarm, a fill light, and a second distribution box; the road crossing video image is collected by the second camera distributed at the road crossing, and the collected road crossing video image is sent to the server through the second switch; the audible and visual alarm is connected to the alarm output interface of the second camera; the second distribution box is electrically connected to the second camera, the second switch, the road crossing barrier, the audible and visual alarm, and the fill light; The perimeter intrusion alarm module includes: a third camera, a third switch, and a third distribution box distributed at the perimeter of the factory area; the third camera distributed at the perimeter of the factory area collects the video image of the perimeter of the factory area, and the collected video image of the perimeter of the factory area is sent to the server through the third switch; the third distribution box is electrically connected to the third camera and the third switch; The first distribution box, the second distribution box and the third distribution box are connected to the power grid.
2. The factory intelligent monitoring system according to claim 1 is characterized in that: The second camera, the second switch, the sound and light alarm, the fill light, and the second distribution box are installed on poles.
3. The factory intelligent monitoring system according to claim 2 is characterized in that: A pole is set on each side of each crossing; two second cameras are installed on each pole, one of the two second cameras is responsible for monitoring the crossing area, and the other is responsible for monitoring the direction of incoming and outgoing trains.
4. The factory intelligent monitoring system according to claim 3 is characterized in that: A signal light is also installed on each vertical pole, and the signal light is electrically connected to the second distribution box installed on the vertical pole.
5. The factory intelligent monitoring system according to claim 3 is characterized in that: The number of fill lights matches the number of second cameras, and the illumination angle of each fill light after installation is consistent with the shooting angle of the corresponding second camera.
6. The factory intelligent monitoring system according to claim 1 is characterized in that: The first camera, the second camera and the third camera are ball cameras or gun cameras.
7. The factory intelligent monitoring system according to claim 1 is characterized in that: There are two level crossing guardrail machines, installed on both sides of the level crossing.
8. The factory intelligent monitoring system according to claim 1 is characterized in that: The server includes: a data server, an image server, a video server, a storage server, and a road crossing remote control device.