Aluminum Tubes,Aluminum Square Tube,Aluminum Round Tubes,Industrial Aluminum Tube Guangdong Jihua Aluminium Co., LTD , https://www.jihua-alu.com
At present, the monitoring of equipment status such as video loss, encoder disconnection, server failure, and hard disk failure is relatively easy to implement, and it is difficult to monitor the quality of some soft faults such as video signal quality, and the system caused by the video quality. Soft faults are the main component of system faults, and the normal use of the system can also cause adverse effects.
The large-scale video surveillance system's image quality diagnostic platform can better solve this problem. It can monitor the signal quality of one image in 3-5 seconds, including signal interference, over-white, black, blurred focus, and images. Freezing and other quality issues can be monitored and alarmed, and different monitoring criteria can be set for different time periods, such as lowering the threshold for over-black monitoring at night and increasing thresholds for over-black monitoring during the day. We use video quality analysis and monitoring algorithm to develop a dedicated analysis server to continuously monitor and monitor the quality of video signals. If there are any video quality problems, a warning message will be displayed on the client interface, or the relevant fault information can be queried through the WEB website. Unified management.
Situation Analysis
According to the Administrative Regulations for the Use of Surveillance Image Resources of a Municipality, in order to strengthen the daily management of surveillance image resources, it is necessary to “establish a routine inspection and maintenance system for surveillance image resources to eliminate problems in a timely manner to ensure the safe operation of surveillance image resources.†In the construction, operation, and maintenance of the city's public safety image resource system, a city has specifically formulated the "Measures for the Construction and Operational Maintenance of a Public Image Resource System in a Certain City", and the construction and operation maintenance of public safety image resource systems and the protection of personnel expenses, etc. Work day-to-day assessments, year-end assessments, and rankings based on assessment scores. This has promoted the establishment of a video surveillance system and the deployment of surveillance personnel in certain aspects. However, in the face of numerous monitoring points, it is difficult to take into account the traditional artificial “eye†inspection method by relying on more than a dozen monitoring screens and several on-duty personnel of the branch monitoring center.
Manual maintenance work is not only time-consuming and labor-intensive, but also it is not effective. Video signals often cannot be detected by maintenance personnel in time after they have experienced different common failures. Some video cameras have either no video data stored or even video data. It is also of poor quality and it is impossible to obtain valuable information content from it;
Due to the limited number of display screens, maintenance personnel often monitor multiple cameras at the same time on one monitor screen or randomly select the camera display, causing some monitoring points to be missed or ignored;
Maintenance personnel have certain instability, arbitrariness, and limitations. People's attention is limited and they are easily fatigued. They can be disturbed by other things, making such manual inspection results unobjective.
Therefore, how to ensure the perfect work of large-scale cameras, video image quality diagnosis system based on video analysis technology provides a good solution to this problem.
The goal of building
This program applies advanced machine learning and computer vision technology to simulate human visual systems. The common camera is a snowflake, scrolling, blurring, color cast, picture freeze, gain imbalance, and out-of-control of the PTZ cameras that appear in front of a city's public safety image resource front camera. Faults and malicious occultation and destruction of monitoring equipment illegal actions to make accurate judgments, and automatically record all test results, generate reports. Users can query the terminal's statistical query history information through the Internet, and statistical analysis of the number of failures and failure rate according to different attributes such as the branch where the camera is located or the police station, brand, type of failure, severity of failure, etc., in order to easily maintain the city public security image resource system. .
Technical route
The video faults are divided into eight types: missing video signal, abnormal video clarity, abnormal video brightness, video noise, video snowflake, video color cast, picture freeze, and PTZ motion out of control. Two kinds of faults, video signal loss and picture freeze, can be concluded by artificially designed method based on video image comparison; PTZ motion out of control is caused by the fault detection system sending out motion instructions, and then through the analysis of the video image motion to detect whether There are faults; for the other five kinds of faults, it is difficult to detect them manually by setting up rules. This requires the machine to simulate the human visual reaction and detect whether the video is faulty.
The video quality diagnosis system adopts video image analysis method to detect various video common faults in the monitoring system. For these five different types of video failures, we designed five different machine learning-based detectors. Each detector is responsible for analyzing whether a video has a certain type of failure and the severity of the failure.
In the actual operation of the video surveillance system to extract a large number of video clips, including normal video and video of various failures, to form training samples, and to simulate human visual characteristics, for a variety of fault types extracted a large number of video image feature parameters for training Get a detector that diagnoses different faults. In the analysis stage, a fixed-length video that needs to be analyzed is obtained. According to the detection item of the video set by the user, different fault detectors are used to extract corresponding video image features and then input to the trained fault detection model. You can get the fault evaluation results for the video.
The video faults that these five detectors can detect are as follows:
· Anomaly detection of video sharpness: An abnormality in focusing occurs; it is blocked by accidental foreign matter (such as catkins); it is artificially blinded;
· Video brightness anomaly detection: too high brightness; gain control disorders (flickering and darkening);
Video noise detection: composite noise of noise and color band rolling;
·Video snowflake detection: snowflake interference; folding interference;
· Video color cast detection: yellow and purple ribbon color cast; simple green color cast; natural green scene, no false positives.
system structure
Regarding the front-end analysis function of the video image quality diagnosis system, the core technology basically determines the fault diagnosis item according to the common fault types of the video. Judging whether this system is good or bad depends on whether the system can effectively integrate with management applications.
The system consists of a video image quality front-end diagnosis system and a video image quality diagnosis back-end comprehensive management system. Among them, the video image quality front-end diagnostic system is responsible for video quality analysis, setting up a diagnostic plan, switching videos one by one according to the list of cameras and inspection items in the diagnostic plan, analyzing and diagnosing the quality of the video, and finally outputting a message to the back-end integrated management system. The quality of the camera's diagnostic results and a video shot.
Video image quality diagnosis back-end comprehensive management system is responsible for providing camera management, camera area management, brand management, user management, log management and other system configuration and management functions, responsible for receiving the video image quality front-end diagnostic system analysis results and stored in the database At the same time, it is responsible for providing WEB-based history query and statistical analysis functions.
The video diagnostics integrated management system provides query and statistical analysis functions based on historical detection records of Web services. Users can access the query websites by typing the corresponding URL in any web browser on any other host that can access the integrated management server. Perform historical queries, statistical analysis, and system management.
System construction benefits
1. Save a lot of manpower and material resources through system construction
The application of the video image quality diagnosis system will greatly relieve the monitoring personnel from the daily inspection of the surveillance points in the jurisdiction, and also avoid the mental fatigue caused by rapid switching of the screen during the inspection.
2. Assist operation and maintenance assessment, greatly improve the efficiency of system security monitoring
The application of video quality diagnosis system provides data and picture support. The system can detect front-end equipment failures in time, and automatically generate various types of reports to show the degree of failure, provide detailed front-end camera operating status reports, and can export and print. The earth facilitates the maintenance and management of large-scale video surveillance systems. The data and pictures obtained by the system can be targeted to the maintenance and update of the equipment through timely feedback from the front-end equipment manufacturers, merchants, or operators. This will ensure maximum availability of the monitoring system and improve the operating efficiency of the monitoring system.
3, avoid redundant construction, save information funds
During the construction of the video image quality diagnosis system, full consideration shall be given to the scalability of various types of businesses in the future, and sufficient interfaces and standards shall be made available; the system shall have high flexibility and be easily interconnected with other systems, and be adapted to future upgrades. Introduce new technologies.
4, to achieve the integration and sharing of information resources
The video image quality diagnosis system supports remote multi-user access to query processing result information, and implements rights management measures to maximize the satisfaction of timely and reliable query of regional monitoring systems of the municipal office and regional branches, and to obtain relevant information. Processing data within the area. At the same time, the platform has set a message board and mailbox function on the WEB interface, which facilitates the timely communication and exchange among relevant management personnel in the city bureau, which makes the information resources shared and communicated maximally, realizing resource integration and data sharing at the same time by regional implementation. Specific management goals.
Conclusion
The large-scale video surveillance system image quality diagnosis platform is a typical application of video analysis technology in the operation and maintenance of the safe city monitoring system. With the increasing number of video surveillance equipment today, it helps users to quickly control the front-end equipment operation and easily maintain a large-scale security system.
System structure of image quality diagnosis platform for large-scale video surveillance system
Intelligent device management can deploy multiple device management servers according to the network topology to perform real-time inspections of devices in the sub-area. This can greatly improve the maintenance efficiency of the system. When the device fails, it should not exceed 10 minutes. The time was monitored and alarmed.