In recent years, there have been many applications of face recognition technology in the field of identity authentication, such as: Alipay, China Merchants Bank's withdrawal, pension collection, etc., but in terms of preventing counterfeiting, authentication security, etc., it is still a comparison. The subject that needs to be further solved, especially in the face detection technology on the mobile terminal. Relying on leading analysis algorithms and technical strength, face live detection can process and return results extremely quickly, meeting the business needs of all walks of life, whether it is financial platforms, social software or various communities, providing customers with reliable authentication services.
Face recognition is a kind of biometric identification technology based on human facial feature information. Use a camera or camera to collect images or video streams containing faces, and automatically detect and track faces in the images, and then perform a series of face-related technologies on the detected faces, usually also called portrait recognition, face recognition.
Today, face recognition has entered all aspects of our lives. Picking up the mobile phone to scan the face to pay bills, scan the face to complete the attendance, and check the hotel with the face has facilitated our life. An indispensable technology in face recognition is face liveness detection, that is, AI not only needs to determine that this is "you", but also needs to determine that this is "the real, living you".
Does it sound weird? Humans can distinguish between living things and objects through vision alone, but machines need to learn to make this detection. Just as you need to complete the recognition by brushing your face, after turning on the phone camera, the machine will detect whether you are "live you". If what appears in front of the camera is a photo, or you in a video, or a human skin mask made of your face, the machine needs to make its own judgment and make sure that the "other you" doesn't pass him. identification. This is the meaning of face liveness detection.
Face live detection is mainly carried out by identifying the physiological information on the living body. It uses the physiological information as a life feature to distinguish biometric features forged with non-living substances such as photos, silica gel, and plastic. In order to ensure that you are the "live you", face liveness detection usually includes several identification steps, such as eye-blink identification: for application systems that can require users to cooperate, the user is required to blink once or twice, and the face recognition system will automatically determine according to The obtained changes in the opening and closing state of the eyes can be used to distinguish photos and faces; or mouth opening and closing discrimination: similar to blinking discrimination, users are required to open and close their mouths once or twice, and the face recognition system distinguishes between photos and faces. real face.
Considering that once the fake face attack is successful, it is very likely to cause heavy losses to users, so it is bound to develop a reliable and efficient face detection technology for the existing face recognition system.
Based on anti-fraud solutions in face recognition scenarios, face live detection technology can effectively block various types of attacks such as PS face swaps, videos, 3D face models, and high-definition portrait photos. Verify that the front-end living body is valid through action instructions, and then obtain a photo of the operator himself, and send this photo to the server side, and the machine will perform the post-test of the living body to prevent attack and fraud and improve security.
Technical communication: 18701686857 QQ: 283870550