What can be done when facial recognition reaches the “Sky Eye” level?

　　In the movie "Fast and Furious", a super hacker program "Sky Eye" system is designed. This program can integrate all the data collection and monitoring equipment in the world. It can collect a large amount of data through mobile phone audio, monitoring cameras and other means, and then use big data and face recognition technology to find the target in a short time . Face recognition has become the key to searching for the target person. In real life, can urban monitoring reach the level of "Sky Eye"? We can start with the actual application of face recognition in current networked monitoring ! Let's follow the security electronics editor to learn about the relevant content.

　　Artistic creation comes from real life. Although the artificial intelligence that makes the protagonist in the movie have nowhere to escape has been artistically processed, it is not completely traceable in existing scientific and technological means. When a system can call up information from various aspects including public transportation, urban electricity, electronic monitoring, banking systems, police systems, etc. at any time, and analyze and integrate all networked information including personal medical records, criminal records, and even online browsing content and input habits, these data can be used to accurately predict people's whereabouts.

What can be done when facial recognition reaches the “Sky Eye” level?

　　In the field of security, face recognition technology is widely used. With the improvement of urban monitoring systems, smart cameras are installed in key areas such as stations, shopping malls, and roads. By performing structured analysis on the shooting targets, valuable information is obtained, which plays an important role in the public security organs' investigation of cases, maintenance of public security, and protection of large-scale events. To realize the practical application of face recognition, three links are indispensable.

　　Front-end intelligent collection

　　At present, some security manufacturers in the industry have launched front-end smart cameras, such as Keda's perception camera series. Generally, smart cameras include three series: vehicle checkpoints, personnel checkpoints, and structured feature analysis. They can identify the content in the monitoring screen based on the video, and perform semantic descriptions and capture the best pictures, and then analyze them through the back-end cloud computing platform to replace us to think and judge. Through the deployment of three cameras according to local conditions, people and vehicles in the monitoring field of view are detected and captured, and main features such as travel direction, clothing color, license plate number, and body color are identified.

　　Personnel bayonet camera

　　In addition to basic feature information such as color and direction, many key places also need to accurately identify and capture photos of people's faces in order to carry out in-depth applications such as face recognition. Such places are very suitable for the deployment of personnel bayonet cameras. Personnel bayonet cameras are specifically for people. Similarly, although feature analysis cameras can also recognize people, they are mainly used in scenes with a large field of view, and it is difficult to accurately capture faces and identify richer personnel feature information. Personnel bayonet cameras can accurately capture the best face photos through cross-sectional videos with a small field of view. Moreover, through patented technology, personnel bayonet cameras can also capture the entire outline of a person. Even if the person is facing away from the camera, the outline of the head and the whole body can be accurately captured. At the same time, the direction and speed of movement of the person can be identified.

　　Perception cameras with perception capabilities are equivalent to visual sensors. The massive amount of information perceived by a large number of cameras enters the big data and cloud computing platforms, allowing us not only to identify content and make judgments from a single camera, but also to make in-depth analysis and mining from massive monitoring data, thus having a profound impact on social management.

　　It provides associated storage of real-time captured head-and-shoulders/full-person photos of people, can perform real-time face comparison, gender, age, and ethnic attribute analysis, and extract the color and texture of pedestrians' upper and lower clothes; supports captured query, personnel search and face search in historical pictures; supports personnel entry, personnel deployment, personnel marking, and alarm query; supports one-click deployment of mobile terminals, and real-time push of alarm information to mobile terminals, etc., to help public security investigators quickly screen personnel targets.

　　Structured analysis system

　　In view of the large workload of checking targets in massive videos in the construction of safe cities, and the difficulty in finding targets quickly due to the different scenes of massive videos, the structured analysis system can conduct secondary analysis specifically for massive images and videos, and is applicable to various media sources with multiple scenes and large differences. It supports real-time analysis of multiple types of front-ends connected; supports secondary analysis such as target detection, attribute analysis, feature extraction, etc. for offline images and videos; and supports distributed deployment and expansion.

　　Video surveillance system

　　The image intelligence system for post-criminal investigation video analysis, management and application is in line with the criminal investigation business process of the public security, and can help criminal investigators focus on the target efficiently and accurately, and perfectly solve various problems in the application of public security criminal investigation videos. The system extracts key information from the video through intelligent analysis technology, stores it in the database in the form of structured data, and supports target retrieval and filtering through attribute conditions such as color and type, tripwires, regions and other intelligent rules, quickly locates key content, and greatly shortens the time to find clues.

　　Image processing system

　　It performs professional technical processing on various low-quality images (videos) to improve image quality, highlight image details, and provide deduction information; it helps users obtain image (video) information of interest, such as license plate numbers, facial information, text information, environmental information, etc., provides clues for cases, clarifies the direction of investigation, and narrows the scope of investigation. It is an essential police equipment for video investigators in the public security industry.

　　Cloud storage construction

　　With the construction of high-definition intelligent front-ends, comprehensive integration and the rapid growth of high-definition video data, video resources and related data are formed and gradually concentrated in large quantities, and the demand for rapid analysis of these resources and data and the requirements for image information exchange and processing efficiency are becoming higher and higher. At the same time, the construction and operation and maintenance costs of traditional data centers continue to rise. It is difficult to meet the above requirements simply by superimposing equipment and improving the performance of a single device. For this reason, many security companies are strengthening the construction of cloud platforms. Deploying video surveillance application systems through cloud computing system architecture will have advantages that video surveillance application systems under traditional architectures do not have: the system platform integrates more comprehensive video image-related information, and information analysis is faster and more accurate; video surveillance comprehensive applications are more efficient and flexible; the entire business system is smarter and more powerful.

　　At present, facial recognition technology has been deeply used in public security and criminal investigation. Although the "Sky Eye" in real life is not as pervasive as in the movies, under the premise of not infringing on citizens' privacy rights and not violating laws and social morals, the existing facial recognition technology is undoubtedly of great significance to maintaining social security and protecting the stability and harmony of residents' lives.

    The above is an introduction to the topic of security electronics: What can facial recognition do when it reaches the "Sky Eye" level? If you want to know more about this topic, please pay more attention to eeworld. eeworld Electronic Engineering will provide you with more complete, detailed and updated information.