Biometrics technology matures: "Brush your face to eat" is no longer a dream
"Brush face eating" "Brush face social" "Brush face pass" in the near future or become a "new normal" of people's lives. At the just concluded 2017 Mobile World Congress in Barcelona, ​​Iris Recognition technology and dual-system DATA Cloud Smart Control Terminal caught the eye of the scene and opened the door to the emergence of iris recognition products. In the meantime, face recognition that belongs to the non-contact biometrics technology with iris recognition is also entering the public eye. In late February of this year, the US authoritative magazine Technology Review released the 2017 top ten breakthrough technologies in the world, among which Chinese technology companies Face ++, Baidu and Alibaba were selected by virtue of their brush face payment technology. When painting a label for technology availability, brush face payments are labeled "Now," which means "brush face eating" "brush face social" "brush face access" in the near future or become people's lives The "new normal." "Brush Face" "Rainbow consciousness" high voice Li Zijun, director of the Chinese Academy of Sciences Institute of Biometrics and Security Technology, told China Science News reporter that non-contact biometrics is a technology that terminal devices can collect identity information without direct contact with users. Iris recognition, people Face recognition, voiceprint recognition, retinal recognition, etc. belong to this type. And voiceprint recognition on the sound environment is higher, the retinal recognition of vision correction may cause inconvenience to the crowd, the industry generally believe that face recognition and iris recognition applications more extensive and convenient, its commercialization is more worthwhile look forward to. Iris recognition has come into everyone's daily life. At present, more than 50 airports all over the world have set up systems that allow passengers to pass through only iris recognition. Singapore also started using "iris scans" this year as a means of identifying travelers and Singaporean citizens in addition to photos and fingerprints. Iris recognition is also playing an active role in the financial field. In order to avoid that the database of "username + password" in the traditional bank is stolen by unscrupulous staff, some banks began to increase the iris recognition verification that only when the iris scanning information and the account information completely match, You can get authorization for the withdrawal of funds, the overall level of security upgrades. Compared with fingerprints and digital passwords, iris recognition is more secure in Internet payment applications, which also makes the iris scanning system have been favored by Samsung and other brands of mobile phones. According to Qiu Xianchao, deputy general manager of Beijing Zhongke Hongba Technology Co., Ltd., "iris mobile phone" will become an indispensable "intelligent organ" for people in the future. However, the public's expectation of face recognition even more than the iris recognition: "brush face eating" "brush face pass" really can be achieved? "Now in Baidu Science and Technology Park, employees will be able to pass the 'Swipe Face Gate' in 1 second, without a work card, without having to wait." Lin Yuanqing, director of Baidu Depth Lab, described Baidu Face Recognition Application of Technology: "The next application scenario is likely to be Baidu canteen - Baidu employees can go to the canteen to bid farewell to the work card, tens of thousands of people will be the first to enjoy the 'brush face' convenience. The pursuit of more natural AI interaction Face recognition has two different scenes, one is 1: 1 face comparison, one is 1: N face recognition. The former common in banks, public security system, some applications, the general need to submit personal identification information and facial features, and then the system ID card and submit the information to be compared 1: 1; the latter has also been in criminal investigation and other fields However, due to the high difficulty of 1: N recognition accuracy, the public security system has traditionally been used to assist in the investigation. It is also "not sure" about the recognition result and "can not be locked" to the suspect. Baidu "brush face" technology choice is 1: N accurate identification. Lin Yuanqing revealed that Baidu has been applying Force 1: N recognition since 2016 and has made great technological breakthroughs. "It has become possible to make 1: N face recognition very precise." Baidu previously announced that Baidu brain face recognition monitoring accuracy has reached 99.7%. "If the accuracy is not enough, in the application of brush face gate can cause Baidu employees were stopped at the door, this situation can not be tolerated." Lin Yuanqing told reporters that Baidu face recognition accuracy of two An indicator: the wrong pass rate and pass rate. "Through rate is easy to understand, refers to the correct rate of matching.Error pass rate means that 'people who are not Baidu easily let go', we are close to 0 this indicator; pass rate is above 99%." In fact 1: N precise recognition to achieve high precision recognition is not easy. This also makes scholars such as Li Ziqing doubt: "Is it really as high as what is advertised?" "1: N is the most difficult to achieve high-precision identification." Lin Yuanqing also admitted that, in particular, as the database increases - the current gates Baidu and future Baidu canteen database for 20,000 to 50,000 people - The accuracy and security of the proposed poses a more serious challenge. "But that's just Baidu's cuisine." Lin Yuanqing said that only achieving 1: N precision identification truly brings about revolutionary changes to the original face recognition methods and makes the interaction in the AI ​​era more intelligent , More convenient. "We must do this because we hope the future AI interface will be very natural and seamless to our life." Quietly into the 2.0 era In fact, when face recognition was first applied, "accuracy" has been the bottleneck in deepening the application. In particular, the initial two-dimensional face recognition does not match the "face" with the "person". "Face tracking recognition technology has entered the 2.0 era." Artificial intelligence technology company ULSee founder Ye Zhou told the "China Science News" reporter: "The traditional face recognition technology is mainly based on visible image face recognition, but this approach Has insurmountable flaws, especially when the ambient light changes, the recognition effect will drop sharply, and there are other issues such as side faces, large positive differences and fraudulent use of photo images. " A biometric technology company's product manager told reporters, "Now, live detection replaced static image recognition, infrared light as a means of collecting facial features to make up for the lack of visible light, live masks, photos, mobile video can not 'Muddle through'." Graphite Part,Other Graphite Parts,High Purity Graphite Part,High Strength Graphite Part CARBONS TECH&CREATION(HENAN) INDUSTRY CO.,LTD , https://www.kymoldedgraphite.com