Among the biometric technologies, face recognition has unparalleled strengths. The video analysis intelligent alarm system based on face recognition can capture the face, human form and other characteristics at any time in different care places for the escaping population, suspect population, controlled population, and out-of-control population. The basic characteristics of the archives in the library are compared, and the identity of the object being guarded is implemented, and the warning is given at any time.
1. Rapid development of biometric technology
Biological identification can be roughly divided into two categories: physical characteristics and behavioral characteristics. The basic characteristics of the body as well as: fingerprint recognition, veins, palms, retina, iris, body odor, face, and even blood vessels, DNA, bones and other characteristics of the path and: signature, voice, walking gait.
At present, biometrics technology is becoming more and more mature, and it has been well utilized in various economic and social industries. It has obvious advantages in dealing with security difficulties and improving task efficiency. In the coming days, the integration and diversification of multiple biometric technologies over time will promote the popularization of products and solutions corresponding to China's multiple biometric technologies, and will actively promote market demand and transform into cross-biotechnology. Emerging industries in the security, IT and other industries.
Integrating bio-resolution technology into the Internet video watchdog platform, integrating the resolution algorithm into the intelligent analysis of video watchdog, will help the accurate analysis of intelligent profiling, and at the same time, rely on high-definition video surveillance technology to drive biological resolution from near The distance is identified by long-distance identification, from static resolution to dynamic identification, from small groups to large groups.
2, face recognition is widely used in intelligent profiling
The intelligent video care system uses networked, intelligent video watchers and image processing, computer vision, and artificial intelligence to presuppose conditions. The video is deployed in the care system to intelligently analyze the video content at any time. The abnormal information data in the caretaker image was found, and an early warning was issued to assist the number of people in charge to handle the crisis. At present, intelligent analysis technology has behavior analysis, basic feature recognition, video diagnosis, and classification calculation.
Through several years of rapid development, the application of video intelligence analysis has a high core density in advanced markets such as the military, political, judicial, financial, transportation, surveillance, nuclear power plants, airports and ports, and then to industrial facilities, education, medical care, etc. The mid-end market has now penetrated into residential markets such as residential areas, retail stores, warehouse management, and logistics. It can be said that the use of video intelligence analysis in the domestic intelligent security has been a hundred flowers.
The video analysis intelligent alarm system based on face recognition can capture the face, human form and other basics in different guarding places for the number of people who are escaping, the number of suspected population, the number of controlled population, and the number of people out of control. The characteristics, after comparison with the basic characteristics of the archives in the template library, realize the identity of the object being watched, and timely warning. The front-end equipment of the system realizes the early warning function of the suspicious target. When it is found that there are suspicious targets within the scope of the guard screen, the high-definition picture of the suspicious target is immediately sent to the center of gravity service platform, and the time generated by the target is marked at the same time, which is convenient for the next time. Check the video in one step.
3, face recognition implant video surveillance technology difficulties
With the tangible development of face recognition technology, face recognition technology will realize the deepening of the digital watch system and become a further application field of face recognition technology. According to statistics, only in the domestic mainland, in the next few years is expected to transform adult sales over 10 billion, and in the next decade is expected to transform into a market scale of annual sales of over 100 billion. Along with the in-depth development of China's peaceful cities and smart city projects, the high-definition of urban guards has become popular, and the total number of cameras has risen on a large scale, which has greatly reduced the face recognition in the extraction of information data and promoted the quality of face recognition. . The connection between face recognition and guard technology has been re-used in the judicial system, which is in line with the urgent need of the public security organ system for functions such as city custody, fugitive pursuit, blacklist investigation, etc., which will be a good helper for the popularization of face recognition technology.
Face recognition is divided into three categories: identification, authentication, and supervision. The identification is a process of comparing the face image of an unknown person with a face image of a known identity in the database to confirm the identity of the unknown person. It is a one-to-many face recognition authentication that is the face image of the person who is now recognized. Aligning with the face image in the database that is claimed by someone else to authenticate the right or wrong order of the identified person, in the case of a one-to-one face recognition monitoring target, to treat an unknown identity The sample to be tested, the face recognition system detects whether it is on the supervision roster, and belongs to one-to-many face recognition.
In the identification, it is the most difficult to monitor the target person. The key reason depends on the discriminative mode of involuntary and the difference of the place. The changes in posture, illumination, age and face resolution seriously affect people. Face recognition rate. Many research companies are always studying how to apply face recognition technology to guards. At this moment, the first thing that should be considered is still taking a good face image. In addition to the algorithmic efforts, from the beginning of the system utilization, the key needs to get the following troubles: system structure difficulties, face resolution problems, gesture troubles, lighting troubles and age difficulties.
English 
简体中文