Face recognition method and device for glasses shielding

Abstract

The invention provides a face recognition method and device for glasses shielding. The method comprises the steps of obtaining a face image and an eye region image of a user to be subjected to face recognition; obtaining face features according to the face image and obtaining eye features based on an attention mechanism idea according to the eye region image; obtaining eye residual features according to the eye features and a preset residual function model; fusing the human face features and the eye residual features to obtain fused features of the human face superimposed eyes; inputting the fusion features of the human face superimposed eyes into a preset human face recognition model to obtain a human face image recognition result; wherein the preset face recognition model is obtained bytaking the fusion features of the face superposition eyes of the training sample as input and taking the correct recognition result of the training sample as output for pre-training. According to theinvention, the recognition accuracy of the face image wearing glasses is improved, and data inclination is avoided.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a face recognition method and device for spectacles. Background technique [0002] With the rapid development of deep learning technology and its excellent performance in face recognition applications, face recognition technology has been applied in many fields such as public security, government, finance, education, and medical care. The application of face recognition technology not only reduces the intensity of manual work, but also effectively improves work efficiency. However, the current face recognition effect is affected by many factors, among which glasses, especially wide-frame glasses, will greatly reduce the accuracy of face recognition. [0003] Aiming at the problem of glasses occlusion, in the prior art, one method is the deep learning face recognition method based on convolutional neural network. Face recognition accuracy under the occlusion of glasses; anothe...

AI Technical Summary

Problems solved by technology

This method will change the face image features to a certain extent, resulting in a certain recognition deviation, and has high requirements for image alignment and high computational complexity, which is difficult to meet the actual application requirements.

Images