Face living body detection and recognition system

Abstract

The invention relates to the technical field of image recognition, in particular to a face living body detection and recognition system, which is characterized in that a power supply module is connected with a light source, an optical imaging system, a data acquisition module, a CPU, an electromechanical control module and a display module and is used for supplying power to connected part units; the light source comprises visible light, infrared light and ultraviolet light sources; the optical imaging system comprises a multiband optical imaging system, an infrared imaging system and a thermal imaging system; the data acquisition module receives image data of the optical imaging system and transmits the image data to the CPU; an image processing module is arranged in the CPU and is used for analyzing and judging after processing image data, and finally displaying the image data on a display module; the electromechanical control module is used for controlling operation of the light source and the optical imaging system. The face recognition method has the beneficial effects that better recognition speed, recognition accuracy and higher safety are realized, the capability of comprehensively resisting most face fraud attacks is realized, and the robustness and safety of face recognition can be improved.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a human face living body detection and recognition system. Background technique [0002] With the rapid development of computer science and electronic technology, face recognition technology is widely used in public security, place entry and exit, information processing and other fields, face recognition is becoming the second largest biometric authentication method in the world after fingerprints, " The era of face brushing has officially arrived. The main purpose of face recognition research is to improve the performance of verification and recognition tasks, but the application of face recognition technology is not flawless. [0003] Compared with 2D recognition, the 3D image itself contains the spatial shape information of the face and is less affected by external factors. [0004] Therefore, the current shortcomings of face recognition are: the deep learning meth...

AI Technical Summary

Problems solved by technology

[0004] Therefore, the current shortcomings of face recognition are: the deep learning method based on the depth information of the face, the model training is difficult, the sample labeling is difficult, and the existing samples are few; the amount of calculation is large, and it is difficult to meet the real-time requirements; High; the dynamic recognition method that requires the active cooperation of the subject to be detected has a response time difference when using random command actions for recognition; face liveness detection is performed based on features such as face texture, color, non-rigid object deformation, and material. The current scheme is easy Affected by factors such as light and environment, there is a certain probability of missed judgments (non-living bodies are judged as living bodies) and misjudgments (living bodies are misjudged as non-living bodies)

[0005] In general, the current face recognition schemes mainly adopt visible light or infrared imaging methods, and there is no anti-fraud face recognition method is mainly aimed at one or several face attack methods, which lacks robustness and needs multiple means of cross-fusion protection and identification

Images