Biometrics Joins with Mobile Internet for Face Recognition

     According to relevant media reports, PayPal (paypal), the world's largest online payment company, has launched a "face recognition" payment function in the UK. people can easily shop with their smart phones, providing convenience for consumers to spend without cards. When the iPhone5S and the iphone5c were launched earlier, the fingerprint identification of the iPhone5s was enviable. It is reported that Android will also follow Apple's iPhone5S to launch fingerprint sensors. With the continuous development of biometrics market and the rapid development of science and technology, especially in the fields of computers and networks, automatic biometrics has been extensively studied and developed. As the frontier of mobile interconnection, its biometric technology has been applied earlier.

      Face recognition has obvious advantages, seizing the market share of biometrics 

     Data show that the global market demand for biometric products will reach 7.1 billion US dollars in 2010. In the next five years, the comprehensive annual growth rate of biometric equipment will reach 21.3%. However, according to the latest authoritative report of the International Biometric Group (IBG), "Biometric Market and Industry Report 2009-2014", automatic fingerprint identification system (AFIS) and real-time scanning account for 38.3% of all biometric technologies. Fingerprint identification accounts for 28.4% (Automatic Fingerprint Identification System (AFIS) and real-time scanning refer to fingerprint identification systems applied in public security, criminal investigation, airports, security check and other industries; Fingerprint identification refers to other civilian markets for fingerprint identification. ) Face recognition accounted for 11.4%; Iris recognition, voice recognition, vein recognition and palm shape recognition account for 8.0%, 3.0%, 2.4% and 1.8% respectively. However, it is understood that the proportion of face recognition in China is 18%, while the proportion of fingerprint recognition has dropped from 98% to 82%. The above data fully show that face recognition has strong development potential and will replace the development trend of fingerprint recognition market.

    It is understood that face recognition is a biometric identification technology based on human face feature information. Images or video streams containing human faces are collected by a camera or a camera, and human faces are automatically detected and tracked in the images, so as to carry out a series of related technologies on the detected human faces, which are also commonly called portrait recognition and face recognition. The application and prospect of face recognition technology in security identification field of entrance guard system, personal identification field, information security field, social entertainment field, etc. It is expected that face recognition technology will have a very broad market space in the future.

    Compared with other biometric technologies, face recognition has its own advantages. First, face recognition has visibility. This is very consistent with the idea of face recognition, only need to see someone's face, you can know who this person is, and then can be used as follow-up evidence. Fingerprints, irises and other biometrics cannot do this. In addition, face recognition requires little or no human cooperation. Fingerprint and iris basically need people to actively cooperate to complete, but face recognition does not need, which determines that the application scene of face recognition is more extensive. Third, face recognition does not require special equipment support. Today's cameras are everywhere. High-definition cameras all over the street and every smart phone can capture or capture human faces, which determines the unlimited popularity of face recognition applications. Face recognition technology is widely used in criminal investigation, entrance guard system, video surveillance, network application, identity recognition, information security and entertainment applications. However, in the latest market reality, face recognition is also applied to the brush face payment system. The first brush face payment system was born and entered the corresponding testing phase. PayPal (PayPal) has launched a "face recognition" payment function in the UK, which indicates that the application of face recognition technology has entered a new field. 

Comparison of Face Recognition Technology with Fingerprint, Iris and Vein Technology 

   Face recognition system widely uses regional feature analysis algorithm, which integrates computer image processing technology and biostatistics principle. It uses computer image processing technology to extract portrait features from videos and uses biostatistics principle to analyze and establish mathematical models. It has broad development prospects. It is more natural, direct and friendly, and has no aggression on the identified person, so that the identified person does not have any psychological barrier. However, relatively speaking, it also has certain shortcomings, such as: face recognition is easy to be cheated by attacks such as face photos and face videos, and is easy to be affected by changes in user's posture, decorations and illumination, thus reducing the recognition performance.

Fingerprint identification relies on the fingerprint characteristics of human body to carry out identity verification. Its advantages are relatively low cost, small volume and convenient feature collection, but it is available for copying and theft, and its safety performance is relatively low. Iris recognition technology based on iris features for identity verification has high security performance, but the image acquisition equipment is very expensive, the size of the equipment is difficult to be miniaturized, the requirements for recognition distance and glasses state are strict, and it is inconvenient to use. Vein recognition is to obtain personal vein distribution map through vein recognition instrument, extract characteristic value according to special comparison algorithm, and obtain finger, palm and back vein map through infrared CCD camera. This kind of technology is not easy to copy, steal, and has high safety performance. However, it is difficult to miniaturize products and has high manufacturing cost due to changes in age and physiology. At present, domestic fingerprint identification accounts for 82% of the whole biometric identification, showing a downward trend, while face identification accounts for 18%, rising gradually, with the trend of occupying the lost market of fingerprint identification. With the development of network and intelligence, the application of face recognition technology will gradually become popular with its convenient performance in the future, which will be on a par with fingerprint recognition. Other biometrics will occupy a certain proportion in special fields, but their extensibility is not as good as face recognition.

Analysis of Future 

   Trend Biometrics will progress with the continuous development of computer technology. Face recognition will be applied more and more widely and further deepened. According to professional analysis, face recognition will show three trends. First, the trend of networking. Face recognition can solve the problem of identity recognition in the daily life of human society. The results of such identification and authentication will be more and more combined with various industrial applications and shared through the network. Secondly, fingerprint authentication technology still occupies a dominant position in the market due to cost and other reasons. However, with the widespread use of face recognition technology in government projects and the application of face recognition in other large industries, the market share of face recognition is rapidly increasing. Third, with the rapid development of chip processing technology, various small devices and even personal digital terminals now have strong computing and processing capabilities and video acquisition capabilities, and most of them have the need to identify user identities, which has created a good platform for the wide popularization of face recognition technology. Embedded application and chip-based face recognition technology will become a mainstream direction of face recognition technology development in the future. Fourth, the development trend of multi-biometric feature fusion of face recognition and other biometric features; Fifth, face recognition technology will be further standardized; Sixth, the organic combination of face recognition and RFID, the Internet of Things application upgrade.