The research of face recognition technology began in the late 1960s. Since the late 1990s, some commercial face recognition systems have gradually entered the market, but these technologies and systems have a certain distance from the practical, performance and accuracy to be improved.
The so-called face recognition technology, that is, based on human facial features, the input of the face image or video stream to judge, first determine whether there is human face. If there is a human face, the position and size of each face and the location information of each major facial organ are further given. And based on this information, to further extract each person's face implied identity characteristics, and it is compared with the known face to identify each person's face identity. The research of face recognition technology began in the late 1960s. Since the late 1990s, some commercial face recognition systems have gradually entered the market, but these technologies and systems have a certain distance from the practical, performance and accuracy to be improved. After the United States suffered a terrorist attack, this technology attracted wide attention. As the most easily concealed use of recognition technology, face recognition has become one of the most important means of international anti-terrorism and security prevention. Recognition of Face Recognition Technology and Its Application
Face recognition technology is divided into two categories: the anti-terrorism security, investigation and evidence collection, criminal investigation-based military / police face recognition technology, and the company management, business privacy-based commercial face recognition technology. The first type of application is more complex, and the most face recognition technology requirements are very high. The second category, mainly concentrated in the company management, business security and other fields, relatively speaking, the technical difficulty is much lower. In the process of identification, the computer through the relevant software on the image of the video image acquisition, face positioning, face recognition preprocessing, identification and identity search and other related measures to identify the image of the people.
This technique mainly adopts the computer image processing technology and biostatistics principle in one, the use of computer image processing technology from the video to extract the portrait feature points, the use of biostatistics to analyze the principle of the establishment of mathematical models, namely the face feature template. The feature analysis is carried out using the built-in face feature template and the person's face of the person under test, and a similar value is given according to the result of the analysis. Use this value to determine whether it is the same person. This technology has now been widely applied to the security, criminal investigation, anti-terrorism, the company secret management and many other areas.
From the practice, compared with other biometric technology, face recognition technology in terms of reliability, accuracy or other data are the best, so it will become the first choice.
Face recognition technology in the field of the latest results
1. R & D of the most accurate facial recognition software in Hong Kong
Recently, the computer scientists at the Chinese University of Hong Kong have developed the world's most accurate facial recognition software, which in the test has achieved near-perfect accuracy, may bring disruptive innovation to the field of public safety technology. This record software was developed by Prof. Tang Xiaoou of the Chinese University of Hong Kong and his team, which achieved 99.15% recognition accuracy in the test, better than all the similar software at present, even higher than the recognition accuracy of 97.53% of human beings.
In the test, this program can be thousands of photos from the exact identification of the same person's two photos, and will not be light, makeup, camera angle and other factors.
The surveillance camera has become a powerful tool for maintaining public safety through which it can monitor a large number of people and find potential dangers, but it is as difficult to identify the suspect as a very accurate person or to find dangerous behavior as a needle in a haystack. Through continuous improvement, this software will help law enforcement agencies or security agencies in the tens of thousands of people looking for individuals, and also by reading facial expressions early prediction of violence.
2. The UK police test the world's fastest face recognition technology
Britain has more than six million closed-circuit television cameras running at the same time, which makes the UK became the most "concern" of the country. These cameras are used to maintain justice "eyes", but the British police are not satisfied with the status quo. Leicester County today confirmed that it will become the UK's first police station to test NEC's NeoFace face recognition technology, and they hope to use this technology to completely change the way to track criminals, thereby improving efficiency. The software is unique in that it can track thousands of faces simultaneously through a monitor probe and compare it to 90,000 photos in real time, and these actions can be done in seconds.
Conclusion
The application of face recognition technology in recent years, more and more attention in the application of security, before the attitude, light and age, face resolution and other aspects of the impact will no longer exist, the recognition rate of the increase to a wider area , Face recognition technology in the monitoring of intelligent applications will also be more and more attention. Fingerprint access control, face recognition access control or mixed identification terminal more and more received the market pro-Lai.
