Human face living body detection method and device and computer readable storage medium

Abstract

The embodiment of the invention relates to the field of image processing and computer vision, and discloses a human face living body detection method and device, and a computer readable storage medium, and the method comprises the steps: obtaining a near-infrared image and a visible light image which comprise a to-be-detected human face; cutting the near-infrared image and the visible light image to obtain a first face image block of the near-infrared image and a second face image block of the visible light image, the first face image block and the second face image block each comprising at least one face feature; and inputting the first face image block and the second face image block into a preset neural network model for living body detection, and judging whether the to-be-detected face is a living body or not. According to the human face living body detection method and device and the computer readable storage medium provided by the invention, the use experience of a user can be ensured while the accuracy of human face living body detection is improved.

Description

technical field [0001] Embodiments of the present invention relate to the fields of image processing and computer vision, and in particular to a human face liveness detection method, device, and computer-readable storage medium. Background technique [0002] Facial images are the most accessible form of biometric identification for highly accurate facial recognition systems, but are also vulnerable to many different types of fake face attacks. Therefore, liveness detection technology, which aims to determine whether the face image captured by the camera is real, has also emerged. The current human face liveness detection methods include: interactive liveness detection based on monocular camera RGB face images, binocular liveness detection based on binocular camera NIR images and RGB images, the above two detection methods are based on manual selection Features (such as LBP, HoG, SIFT, SURF and DoG) to obtain the feature distribution of living and non-living, so as to distin...

AI Technical Summary

Problems solved by technology

[0003] The inventors found that there are at least the following problems in the prior art: at present, liveness detection based on monocular RGB face images is mainly interactive liveness detection, which requires the user to cooperate with actions such as nodding, opening the mouth, and blinking. Poor; binocular liveness detection based on two kinds of face image information of binocular camera NIR image and RGB image, mainly using manually selected features, and then using SVM (support vector machine) for classification

The artificially selected features are very sensitive to illumination, facial posture, and special-looking faces, while the SVM model cannot capture ambiguous information, and its generalization ability is poor, resulting in low accuracy of face detection

Images