Living body detection method utilizing eye tracing

Abstract

The invention discloses a living body detection method utilizing eye tracing, and aims to solve the technical problem that whether a detected object is a true living body can not be judged accurately due to the fact that a detected person can use methods such as a photo, a video, a waxwork image to cheat a human face detection system in current face recognition and tracing application. The detected object is asked to use eye movement to trace the dynamic moving locus of a randomly selected object at real time in the living body detection method utilizing eye tracing, according to the collected eye image in eye real-time tracing process, the eye movement locus of the detected object is obtained, the obtained eye movement locus is compared with the pre-designed dynamic moving locus of the randomly selected object, so that the authenticity of the human face living body is judged and recognized. The living body detection method utilizing eye tracing has the advantages of being simple, easy to operate and effective. The detected object is asked to use eye movement to trace the pre-designed dynamic moving locus of the supplied random object at real time to recognize the authenticity of the human face living body of the detected object, and accordingly the technical problem that the detected person use the photo, the video, the waxwork image and the like to cheat the human face detection system is effectively resloved.

Description

Technical field: [0001] The invention relates to a method for detection of living human faces, in particular to a method for detecting human faces by using eyeball tracking. It can be widely used in various application fields that need to recognize authenticity through face recognition and liveness detection. Background technique: [0002] At present, in face recognition and tracking applications, liveness detection is usually judged by blinking, that is, prompting the detected person to blink and detecting whether to blink. This method can prevent photo deception, but for recording video, the avatar in video playback It will keep blinking, which will make the detection system unable to accurately judge whether the detected object is a real living body. Therefore, the detected person can deceive the face detection system with photos, videos, wax figures, etc., especially dynamic video recordings are more deceptive, making it difficult for the detection system to identify th...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the current technical problem in face recognition and tracking applications, the detected person can deceive the face detection system with photos, videos, wax figures, etc., and cannot accurately judge whether the detected person is a real living body, Provides a method for liveness detection using eye tracking. The method requires the detected object to track the dynamic trajectory of the randomly selected object in real time by using the eye tracking dynamically, and acquires the captured eye image during the real-time tracking of the eye. Detect the eye movement trajectory of the object, and then compare it with the pre-designed dynamic trajectory of randomly extracted objects to judge and identify the authenticity of the live face

Images