Face recognition method and device based on AI trinocular imaging technology

Abstract

The present application discloses a face recognition method based on AI trinocular imaging technology, which belongs to the technical field of face recognition. Including: acquiring and identifying the infrared temperature feature in the infrared thermal imaging picture; judging whether there is an infrared temperature feature that is not related to the infrared temperature feature function of the face; comparing with the preset database, and identifying the type of occluder; using occlusion The temperature feature function of the object obtains the infrared temperature feature of the occluded area of ​​the face; generates the infrared temperature feature of a virtual face; and performs face recognition according to the infrared temperature feature of the virtual face. In addition, a face recognition device based on AI trinocular imaging technology is disclosed. Based on the traditional binocular face recognition technology, this application adds an infrared biometric recognition sensor, which creatively solves the infrared feature recognition of human face occlusions and the influence of infrared features of occlusions, obtains virtual faces and analyzes virtual face features. Fast recognition greatly improves the accuracy of face recognition.

Description

technical field [0001] This application relates to the technical field of face recognition, in particular to a face recognition method and device based on AI trinocular imaging technology. Background technique [0002] Most of the face recognition systems currently used in the market are face recognition systems based on binocular imaging technology. Although this recognition system can basically solve the problem of living body (photo) recognition, it is difficult for the face to be blocked by hats, masks, scarf fabrics, etc. In this state, the accuracy of face recognition will be greatly reduced, and it will not be able to meet the needs of application scenarios with strict identification requirements and public health and epidemic prevention security. [0003] The biggest defect of the existing technology is that when there is an occluder on the face, it becomes very difficult to accurately identify the identity. In addition, because the binocular recognition technology o...

AI Technical Summary

Problems solved by technology

[0002] Most of the face recognition systems currently used in the market are face recognition systems based on binocular imaging technology. Although this recognition system can basically solve the problem of living body (photo) recognition, it is difficult for the face to be blocked by hats, masks, scarf fabrics, etc. In the state of face recognition, the accuracy of face recognition will be greatly reduced, and it cannot meet the needs of application scenarios with strict identification requirements and public health and epidemic prevention safety.

[0003] The biggest defect of the existing technology is that when there is an occluder on the face, it becomes very difficult to accurately identify the identity. In addition, because the binocular recognition technology only uses two cameras, the recognition distance and range are relatively limited, and the identity recognition is limited to Even more, it is impossible to identify the identity of the person whose face is under the influence of the occlusion, which leads to a significant drop in the accuracy of face biometric database comparison in the state of occlusion.

Images