Face recognition anti-spoofing method, system and terminal based on infrared rays

Abstract

The invention belongs to the technical field of face recognition, and discloses a face recognition anti-spoofing method, a system and a terminal based on infrared rays, and the method comprises the steps: employing a near-infrared camera to obtain a near-infrared face image of a user; carrying out image enhancement on the acquired near-infrared human face image; constructing a face authenticity classifier based on a convolutional neural network; constructing a real and forged face data set under near infrared rays, and training a face authenticity classifier based on a convolutional neural network by using the near infrared face data set; and performing an anti-spoofing test on the face input by the user, if the output of the system is a real face, judging that the user belongs to a real identity, and allowing access to a subsequent face recognition process. And if the output of the classifier is a forged face, a fact that the user performs spoofing attacking is determined, the user isforbidden from accessing a subsequent face recognition process, and alarm processing is performed. According to the method, the defect that a traditional face recognition technology is poor in safetyor insufficient in anti-spoofing capacity is overcome, and compared with other anti-spoofing methods, the method has higher precision and higher robustness.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and in particular relates to an infrared-based face recognition anti-spoofing method, system and terminal. Background technique [0002] At present, the research and development of identification systems are very important for social life and personal daily life. Biometric identification is a technology for personal identification based on the inherent physiological characteristics of the human body (such as hand shape, fingerprints, facial features, iris, retina, etc.) and behavioral characteristics (such as handwriting, voice, gait). Traditionally based on passwords, identity cards, etc., it does not need to be carried, memorized, and will not be lost or damaged. Among them, face recognition technology has attracted more and more social attention in recent years because of its rapid development in theory and practical application. However, face information is easy to leak, and face r...

AI Technical Summary

Problems solved by technology

[0007] (1) The existing face recognition technology often ignores or does not consider face spoofing attacks, and is vulnerable to forgery attacks;

[0008] (2) Many existing anti-spoofing technologies are based on traditional manual design methods, which are complicated to use and not very accurate;

[0009] (3) Existing anti-spoofing technologies have their own special situations, but their versatility and robustness are poor;

[0010] (4) Most of the existing anti-spoofing technologies are not effective against environmental lighting changes, 3D mask deception, etc., and some of them require the cooperation of users, which is not convenient enough

However, the feasibility of using infrared imaging for face anti-spoofing is still unknown and needs to be verified experimentally.

In addition, the infrared face data collected by the infrared camera is often not directly usable, and necessary preprocessing and image enhancement are required. This is also one of the difficulties in anti-spoofing based on infrared faces.

Images