RecordTrend.com is a website that focuses on future technologies, markets and user trends. We are responsible for collecting the latest research data, authority data, industry research and analysis reports. We are committed to becoming a data and report sharing platform for professionals and decision makers. We look forward to working with you to record the development trends of today’s economy, technology, industrial chain and business model.Welcome to follow, comment and bookmark us, and hope to share the future with you, and look forward to your success with our help.
The following is the Research Report on face recognition industry in 2020 From China Institute of electronic technology standardization recommended by recordtrend.com. And this article belongs to the classification: artificial intelligence, research report.
1. Principle of face recognition technology
Today’s mainstream face recognition algorithms mainly include five steps: face detection, face preprocessing, feature extraction, comparison recognition and live identification. Face detection, face preprocessing and feature extraction can be collectively referred to as face view parsing process, that is, to detect faces from videos and images, select appropriate face images through image quality judgment, extract face feature vectors for subsequent comparison recognition; comparison recognition processing can be divided into face verification (1:1) and face recognition (1:n); The biometric algorithm is used to judge whether the face image in face recognition processing is collected from the real human body. In practical application, in addition to the above face recognition algorithm, front-end view collection technology, face data storage technology, application software management technology are also important parts of face recognition technology.
2. Science and technology enterprises
In the field of face recognition technology research, many science and technology enterprises also play a crucial role. Microsoft Asia Research Institute started the research on face recognition technology earlier and published a large number of excellent academic papers. In 2018, the deep learning residual network RESNET proposed by Microsoft Asia Research Institute has been widely recognized in the research field. Apple has conducted in-depth research on face recognition technology, and since 2017, it has introduced the face brushing and unlocking function on its iPhone x mobile phone; NEC is also one of the pioneers of face recognition technology in the world. It has proposed a public safety solution based on face recognition technology for a long time; There are “four little dragons of artificial intelligence” in China. Shangtang, Kuangshi, Yitu, Yuncong and other enterprises have done a lot of work in the field of face, from academic research to industrial practice, and have made new achievements in the fields of complex scenes and large-scale processing; Baidu, Alibaba, Tencent, Ping’an technology, Haikang, Dahua and other traditional domestic technology enterprises have also carried out extensive and in-depth research in the field of face recognition technology, and achieved remarkable technical research results in combination with their original business scenarios.
3. Technical advantages
In different biometric recognition methods, face recognition technology has its own special advantages, so it plays an important role in biometric recognition.
(1) Non intrusive, face recognition can achieve better recognition effect without disturbing people’s normal behavior. As long as you naturally stay in front of the camera for a moment, the user’s identity will be correctly recognized.
(2) Convenience, face recognition acquisition equipment is simple, fast to use. Generally speaking, the common camera can be used for face image acquisition, without special complex equipment. Image acquisition can be completed in a few seconds.
(3) Friendly, the method of identity recognition by face is consistent with human habits, people and machines can use face images for recognition.
(4) Non contact, face image acquisition, users do not need to contact the device directly. In addition, face image can be collected in a long distance. The focal length of the camera equipped with optical zoom lens can be increased to more than 10 times, and the depth of field can be extended beyond 50 meters. The camera can take clear pictures of the distant face image effectively.
(5) Scalability: After face recognition, through the next processing and application of the recognition result data, many practical applications can be expanded, such as access control, face image search, card swiping, illegal person recognition and other fields.
(6) Face recognition has more advantages than fingerprint in this aspect.
(7) The face information recorded by the system is a very important and easy to use clue, which is more conducive to the application of post tracking.
(8) Compared with human body, gait and other features, face features are more discriminative and have lower false alarm rate. The scale of the base database can be applied is much higher. At present, super large scale (billion level) face retrieval can be applied.
4. Technical limitations
Due to the similar face, age, algorithm bias, the diversity of scenes and the easy access of face images, face recognition technology itself also faces some limitations.
(1) Similar faces are more difficult to solve. Twins or very similar faces are prone to recognition errors, but there is no new technology to solve this problem. NIST analysis report points out that in most cases, twins can still distinguish between high and low scores, but they are often above the threshold, and the application effect is poor in the open environment.
(2) algorithm bias. Because the current face recognition algorithms rely on data samples to a large extent, but the face data samples of different groups are different, which leads to the different recognition ability of the algorithm for different regions and different age groups. NIST’s inspection shows that face recognition software has great differences in different regions, races, genders and ages. For example, children, the elderly and other rare race or skin color face recognition rate is relatively low, which needs to be solved.
(3) Face recognition rate is easily affected by many factors. The existing face recognition system can achieve satisfactory results under the condition of user cooperation and ideal acquisition conditions. However, if the user does not cooperate and the acquisition conditions are not ideal, the recognition rate of the existing system will be affected. For example, according to the NIST test report, the error rate of most algorithms will increase by more than one order of magnitude in the case of wearing masks. Factors such as cross age, large angle and so on will also cause varying degrees of decline.
(4) The influence of age change. With the change of age, the facial appearance also changes, especially for teenagers. For different age groups, the recognition rate of face recognition algorithm is also different.
(5) Security issues. Face recognition system information storage will also face the attack of hackers. So data encryption is very important. With the continuous improvement of technology, the security of face recognition technology needs to be strengthened. At the same time, face exposure is higher, and it is easier to achieve passive acquisition than other biometric data. It also means that the data of face information is easier to be stolen, which may not only infringe personal privacy, but also bring property losses. Large scale database leakage will also bring security risks to an ethnic group or a country.
If you want to get the full report, you can contact us by leaving us the comment. If you think the information here might be helpful to others, please actively share it. If you want others to see your attitude towards this report, please actively comment and discuss it. Please stay tuned to us, we will keep updating as much as possible to record future development trends.