Intelligent video analysis becomes a highlight of security technology research and development

    Intelligent video analysis technology (VideoAnalytics) integrates the research results of multiple disciplines. It mainly includes image processing, tracking technology, pattern recognition, software engineering, digital signal processing (DSP) and other fields. After the "911" incident in 2001, the United States greatly increased its investment in security research. Many research institutions and researchers have joined the research and development of security technology, and intelligent video analysis is one of the highlights.

　　There are differences in the implementation methods of intelligent video analysis

　　At present, intelligent analysis technology has been widely used in various security fields. According to the way it is implemented, some industry insiders summarize it into the following types of intelligent analysis.

　　Diagnostic intelligent analysis: Diagnostic intelligent analysis mainly analyzes, judges and alarms common camera failures, video signal interference and video quality degradation such as snow, scrolling, blur, color cast, gain imbalance, PTZ out of control and frozen images. Diagnostic intelligent analysis technology is relatively simple to implement and usually appears in the form of a back-end management platform. It plays a very important role in the daily operation and maintenance of large-scale monitoring projects, especially city-level monitoring. At present, some DSP-based intelligent analysis devices, DVRs and DVSs on the market have this auxiliary function.

　　Identification-based intelligent analysis: This technology tends to analyze and process static scenes. Through core technologies such as image recognition, image comparison, and pattern matching, it can extract and analyze relevant feature information such as people, vehicles, and objects. The main application of vehicle recognition and analysis is license plate recognition technology. After years of development and application, this technology is now very mature. The emergence of fuzzy license plate restoration and recognition technology has made this technology no longer limited to high-definition and has begun to spread to the standard-definition field. License plate recognition technology is widely used in parking lot entrances and exits, highway toll stations, etc., and has developed rapidly in recent years: in conjunction with the traffic electronic card system, license plate recognition technology is widely used to capture vehicle traffic violations, effectively reducing the number of vehicle traffic violations and greatly reducing the occurrence of traffic accidents.

　　Behavioral intelligent analysis: This technology focuses on the analysis and processing of dynamic scenes. Typical functions include: vehicle reversal and related traffic violation detection, zone intrusion detection, wall climbing detection, tripwire crossing detection, item theft detection, road occupation detection and passenger flow statistics. Mobile motion detection (VMD) is the "early intelligence" in this type of intelligent analysis. VMD makes judgments based on the movement changes of pixel blocks in the video screen. It has obvious shortcomings: based on two-dimensional analysis, the false alarm rate is too high, and it is impossible to identify whether the moving pixel blocks are interference or targets.

　　Main forms and features of products

　　At present, intelligent video analysis products are mainly based on general-purpose CPUs such as Intel (servers, industrial computers) or DSPs. Some products are integrated with DVRs, and some products are made into independent modules, which are used by integrators by providing interfaces and developing SDKs. The most integrated products are integrated with cameras and directly output intelligent analysis results.

　　Server-based (industrial computer) systems are usually suitable for deployment in the background of monitoring systems. Because of its relatively open architecture, it can be easily integrated with existing monitoring systems. In addition, the CPU processing power of servers is much higher than that of DSPs, and more complex algorithms can be used. The optimization operations that were previously unique to DSP processors are gradually becoming common on general-purpose processors. Multi-core is the development direction of Intel CPUs, which is very suitable for the needs and development trends of multi-channel image processing and is very helpful in reducing system costs. Intel's speed of launching a new product every two years is also much faster than that of Texas Instruments (TI). The performance of server-based systems can be easily improved with the update of Intel products. In some high-end intelligent video monitoring systems, servers are used more frequently. DSP modular products are usually suitable for deployment in the front end of monitoring systems. It is relatively easy to install and implement, and it is also easy to protect intellectual property rights.

　　The successful installation and use of intelligent video analysis products should usually be directly and effectively supported by experienced integrators or developers in the form of solutions. Compared with traditional monitoring forms, the intelligent video analysis products on the market have made substantial progress in monitoring efficiency and quality, especially in real-time alarm. However, they are still a little far from the standards we expect, and we cannot expect to set up a system and run it normally immediately like installing a surveillance camera. Like any new technology, to achieve good results, engineers who understand the characteristics of intelligent video analysis products must provide full support from solution design to construction and debugging.

　　Development direction of intelligent video analysis applications

　　The application of intelligent analysis industry is diverse. One is the intelligent recognition technology represented by license plate recognition and face recognition, which is mainly used in electronic police, airports, and customs. The other is the behavior analysis technology represented by perimeter defense, crowd counting, automatic tracking, reverse driving, no parking and other rules, which is mainly used in perimeter warning areas, shopping malls, transportation, scenic spot flow statistics, road parking and no-parking, illegal reverse driving, scene tracking and other aspects. For example:

　　1. Dual-machine automatic tracking: intelligent analysis camera plus ordinary fast ball mode. Can be applied to urban alarm emergency plan. Object tracking in emergencies.

　　2. Crowd flow statistics: Count the number of people entering and leaving the selected area, which is applied to the analysis and statistics of customer flow in supermarkets and shopping malls to help merchants formulate corresponding sales strategies. It is applied to scenic spots and subway entrances to provide flow data for personnel control applications.

　　3. Crossing the warning area: By setting up a virtual fence, the perimeter is detected. When a suspicious person or object is found crossing the fence, an alarm is triggered and the alarm signal is uploaded to the monitoring management center. At the same time, the alarm screen can be uploaded to the remote monitoring user through the network. It is used in traffic road crosswalks or zebra crossings, factory key area fences, schools, detention center fences, etc.

　　4. Lost analysis: By drawing an area on the monitoring screen where important items are placed as a warning area, as long as the item leaves the warning area, the alarm rule will be triggered immediately. It is used in key protection areas such as museums, exhibition halls, auctions, gold and silver shops, etc.

　　5. Direction analysis: In actual monitoring, people may be concerned about the direction of pedestrian and vehicle traffic. Direction recognition can be used to determine whether the target is walking or driving illegally. If there is reverse behavior, the target will be automatically locked and an alarm will be issued at the same time. It is used in one-way roads, important entrances and exits, etc.

　　6. Intelligent tracking: The target is locked on a suspicious person or object, and the target's movement trajectory is recorded. At the same time, the camera will follow the target and alarm. It is used in high-end residential areas, restricted areas, confidential areas, important protection areas, etc. It can also be used as a trajectory analysis of the case playback process after the incident, so as to quickly solve the case.

　　The intelligent video analysis system frees security personnel from the complicated and boring task of "staring at the screen", and the equipment completes this part of the work; another is to quickly search for the desired image from the massive video data. Studies have shown that operators will miss 90% of the video information after staring at the TV wall for more than 10 minutes, making this work meaningless. Often the monitoring system becomes the basis for tracing the incident afterwards. However, the intelligent video analysis system turns the post-event evidence collection into active defense, allowing the project to achieve the most effective security at the lowest cost. The intelligent video analysis system has been successfully applied to various industries and has played a pivotal role.

　　Conclusion

　　As people's understanding of intelligent analysis gradually deepens, their understanding of intelligent video analysis has gradually shifted from perceptual to rational. Intelligent monitoring is an important development direction for future monitoring. The construction of safe cities and large-scale network monitoring systems have put forward urgent demands for the application and development of intelligence. At the same time, this will also greatly promote the development of intelligent technologies and products. The video surveillance market is booming and demand continues to increase. It is believed that with the continuous improvement and perfection of intelligent video analysis algorithms, intelligent video analysis products will be more and more widely used in social production and life, and play an increasingly important role in ensuring the harmonious development of society and improving the quality of life.