At present, China’s face recognition technology is mainly used in security and finance. According to incomplete statistics, 72% are used in security and 20% in finance. In the field of security, the personnel access control system accounts for the vast majority of the market, mainly in the form of face access control, face channel, face attendance and other products.

The application advantages of face recognition, such as non-contact, convenience and scalability, make it widely used in many application scenarios. In addition to the large-scale application in personnel access scenarios, the value of face based identity authentication has also been further explored. Some special scenes begin to carry out personalized applications based on faces, such as face running red light system at intersections and face based traffic and settlement applications in unmanned supermarkets. For another example, some government businesses use face recognition to implement the “real person system” management, such as pension withdrawal, real person occupancy of indemnificatory housing, etc. Overall, the acceptance of face recognition in the market exceeded expectations.

Security and scene customization of face recognition

Face recognition belongs to technical prevention, so there are some loopholes. From the perspective of products, there are two major security risks: one is the privacy disclosure of personnel information, and the other is anti-counterfeiting. Privacy disclosure is a security policy vulnerability of the system. The key to filling the vulnerability depends on how the user uses the data; Anti counterfeiting is the standard to verify the ability of front-end recognition terminal equipment and algorithms. Strengthening the construction of face anti-counterfeiting is also the key to promote the landing of face recognition in more application scenarios.

At present, most face recognition products in the market have anti-counterfeiting functions, and there are a variety of anti-counterfeiting methods, such as common 3D structured light, iris detection + face recognition anti-counterfeiting, near infrared + visible light anti-counterfeiting, etc. their technical logic is to detect whether the target object is a living body through one technology or a combination of multiple technologies.

At this stage, the application and promotion of face recognition products has sent a wrong message to the market – the fast recognition speed is a good product. In fact, second level recognition is only one of the standards, as well as the error rate and safety level of the minimum probability. Face recognition access control or access gate series products need to prevent unauthorized personnel from entering. In this scenario, it is also necessary to consider that other people follow the gate.

Therefore, the traffic experience is mainly a combination of products and scenes to form an independent scene solution, which can make the traffic experience smoother. For example, if domestic face recognition products are applied to Europe, America and Africa, how will the recognition effect be? Different scenes have different experience effects. Face recognition products suitable for university dormitory management will have different application effects if they are placed at the gate of kindergartens. Although face recognition has a wide application field, the scene is scattered and fragmented. How to solve the scene fragmentation is a problem that every manufacturer should consider. For example, the scenes of colleges and universities contain many subdivided scenes: gate, canteen, dormitory, library, classroom, scientific research building, Key Laboratory, gymnasium and supermarket. Each scene depends on products and solutions.

Policy and market are favorable, and the face recognition market can be expected in the future

Domestically, the explosive growth of face recognition industry in the past two years is mainly due to the fact that the state attaches great importance to the development of face recognition industry and has issued a number of policies to boost the development of the industry. Since 2015, the state has intensively issued the guidance on remote opening of RMB accounts by banking financial institutions (Draft for comments), which has opened the door for the popularization of face recognition; Since then, laws and regulations such as technical requirements for security video surveillance face recognition system and safety technical requirements for information security technology network face recognition certification system have laid a solid foundation for the popularization of face recognition in the fields of finance, security and medical treatment and cleared away policy obstacles; At the same time, in 2017, artificial intelligence was written into the national government report for the first time. As an important subdivision of artificial intelligence, the national policy support for face recognition is increasing; The three-year action plan for promoting the development of a new generation of artificial intelligence industry (2018-2020), released in December 2017, specifically plans to “by 2020, the effective detection rate of face recognition in complex dynamic scenes will exceed 97% and the correct recognition rate will exceed 90%”.

In the overseas market, the United States has asked its customs and border inspection to start using face recognition and human card verification and customs clearance products, which also means that face recognition technology also has the opportunity and trend of popularization and application in some overseas markets that pay attention to personal privacy, and the popularization and application of overseas face recognition is full of imagination.

Based on the current application status of face recognition in the market, from the perspective of security, product form and application scenario, we can predict: 1. 3D face recognition will be widely used; 2. The face recognition and access control all-in-one machine of Linux system will replace the face recognition and access control all-in-one machine of Android system and become the mainstream; 3. Face recognition products pay more attention to combined application with scene; 4. Face recognition is mixed with other recognition technologies to make up for the application shortcomings of face recognition.

Leave a Reply

Your email address will not be published. Required fields are marked *