Living body detection device and method

Abstract

The invention discloses a living body detection device and method for face recognition. The living body detection device comprises an infrared camera, a coaxial infrared light source, a paraxial infrared light source, a module control and data processor and other necessary accessories, wherein the infrared camera, the coaxial infrared light source and the paraxial infrared light source are fixed according to a specific mode. A living body detection technology based on the device comprises the following steps: according to a security level requirement and a termination condition input by an administrator, generating a binary random security code sequence before each detection; according to the value of each digit of the security code sequence, sequentially controlling the on / off of the coaxial infrared light source and the paraxial infrared light source according to the method disclosed by the invention, collecting the face image of the living body detection object, and further analyzing whether the image has a bright pupil effect or a dark pupil effect; according to the random security code sequence and the detection result of the bright pupil effect or the dark pupil effect of the corresponding image, a living body detection result is obtained through analysis, and whether the living body detection object is a real person or a face prosthesis is judged.

Description

technical field [0001] The present invention mainly relates to the field of identity recognition, in particular to the field of preventing face prosthesis attacks in face recognition. Background technique [0002] In the field of identity recognition, face recognition systems and methods are widely used. However, in practical applications, as a feature that is easily obtained by a third party, a fake face prosthesis may be used to carry out non-personal attacks on the face recognition system. Face prostheses include but are not limited to photos, videos of face recognition subjects, or 3D solid models made based on their facial features. [0003] With the commercialization of the face recognition system, some unattended and automated application scenarios have emerged. How to accurately, automatically and efficiently distinguish whether the object detected by the face recognition system is a real face or a face prosthesis has become an urgent task. issues that need resolvi...

AI Technical Summary

Problems solved by technology

The human eye can produce bright pupil effect or dark pupil effect under specific light, and under the existing technical conditions, the artificial eye that simulates the bright pupil effect and dark pupil effect is expensive, and generally has a large volume, which is difficult to integrate into the human body. face prosthesis

Images