A test device and test method for a multi-eye face detection system

Abstract

The invention discloses a test device for a multi-eye face detection system, comprising: a plurality of projectors respectively corresponding to a plurality of cameras, a background screen for projecting images, and a spectroscopic unit; the projector is placed on the background screen For projection, each camera simultaneously and separately shoots the projection image of the corresponding projector on the background screen to obtain the received image, and makes the received image of each camera the same as the real face image captured; the light splitting unit is used to ensure that each camera only shoots Corresponds to the projected image of the projector. The test method of the present invention is: respectively reconstructing the real face images taken by each camera, respectively projecting the reconstructed images of each camera, and the multi-eye system performs face liveness detection based on the received images of each camera, and according to the The result of face liveness detection realizes the test of multi-eye system. The invention has the advantages of simple operation, easy realization, strong reproducibility, low cost of repeated experiments, and is beneficial to expand the test set for full testing.

Description

A test device and test method for a multi-eye face living body detection system technical field The present invention relates to the technical field of human face liveness detection, especially a kind of detection system for multi-eye human face liveness detection system. Test device and test method. Background technique [0002] Face liveness detection and recognition technology have been widely used in the field of identity authentication, such as access control, payment, Gate and other scenarios, among which, face live detection technology is an important line of defense to prevent counterfeiting and security certification. Continuously improve the living body of the face The security of detection is also the goal that face recognition solution providers continue to improve. At present, the living body detection scheme is also constantly expanding, and the common means are binocular, structured light, etc., that is, the three-dimensional Information for live dete...

AI Technical Summary

Problems solved by technology

[0005] The 3D face model required for the multi-eye liveness detection system test usually needs to use a high-precision third-party scanner to obtain high-precision 3D face data, and then use high-precision 3D printing to make a 3D face mask or model, the production process is cumbersome

The production cost of this kind of 3D face mask or model with high fidelity is usually several thousand yuan or even hundreds of thousands, which is expensive

Due to the cumbersome and expensive production process of this 3D face mask or model, it is difficult to expand the test set, and it is often difficult to fully test using this method

Images