Living body detection method and device and computer device

Abstract

The embodiment of the invention provides a living body detection method and device and a computer device. The method comprises the steps of obtaining a to-be-detected image of a to-be-detected object;performing cutting according to a plurality of scales; obtaining cropped images with different scales; at the moment, since the cropping scales are different, feature information contained in the cropped images with different scales, except the human face area, being different; inputting the plurality of cropped images with different scales into a neural network. According to the method, the feature information of the cropped images of different scales can be processed in a targeted mode, the processing results of the cropped images of different scales are fused for living body detection, thedetection accuracy and efficiency are improved, and attack samples such as human face paper scraps, duplicates, human face video synthesis and masks can be effectively recognized. Moreover, in the whole detection process, the to-be-detected object does not need to cooperate with the interaction action according to the instruction, the operation is simpler and more efficient, the interaction costis reduced, and the user experience is improved.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to a living body detection method, device and computer equipment. Background technique [0002] Liveness detection refers to the computer to determine whether the detected face is a real face or a forged face attack, such as a legitimate user picture, a video shot in advance, etc. In this regard, in current liveness detection applications, the user is usually required to cooperate with action detection, such as performing face actions such as blinking, nodding, and opening mouth according to instructions, and through corresponding action detection, face detection and other operations to determine whether the current user is a real person or an attacker sample. [0003] However, the existing liveness detection method is vulnerable to the attack of forged video combined according to the command action in advance, and takes a long time, which greatly reduces the eff...

AI Technical Summary

Problems solved by technology

[0003] However, the existing liveness detection method is vulnerable to the attack of forged video combined according to the command action in advance, and it takes a long time, which greatly reduces the efficiency of liveness detection

Images