Face identification method for multi-patch multi-channel combined characteristic selection learning on the basis of CNN (Convolutional Neural Network)

Abstract

The invention discloses a face identification method for multi-patch multi-channel combined characteristic selection learning on the basis of a CNN (Convolutional Neural Network), and belongs to the technical field of face identification. The method comprises the following steps that: firstly, segmenting an original face image into a plurality of sub-images, and separating each sub-image into a plurality of channel images; then, constructing a CNN model for each channel image, and inputting a channel image for identification; and then, connecting the plurality of channel neural networks of thesame sub-image to obtain a plurality of sub-image neural networks corresponding to the plurality of sub-images, and then, connecting the plurality of sub-image neural networks to serve as a final model identification result. By use of the method, an existing CNN model is improved and innovated so as to achieve an effect of model optimization and improvement, the face identification ability of theCNN model is more accurate, and a powerful technical guarantee is provided for widely applying the method in fields including daily life, industrial development, scientific research and the like.

Description

Technical field [0001] The invention belongs to the technical field of face recognition, and specifically relates to a face recognition method based on a convolutional neural network. Background technique [0002] In recent years, biometric-based identification technology has been widely used in many scenes of daily life. Among many biometric technologies, face recognition technology has the advantages of non-invasiveness, non-contact, easy operation, etc., and it is easier to collect facial image data. In this way, the application scenarios of face recognition technology in the fields of information security, identity verification, site monitoring, and human-computer interaction have become more extensive. Therefore, in-depth study of face recognition has important theoretical and practical significance for attendance, safety, entertainment and other aspects. [0003] At present, the common face recognition methods mainly include: face recognition methods based on geometric feat...

AI Technical Summary

Problems solved by technology

Although the current face recognition method based on deep learning is better than many traditional algorithms in terms of accuracy, it also has some shortcomings. For example, this method often ignores some local key features of the face, and only Feature learning will be performed on the overall face of the original image, which reduces the accuracy of face recognition to some extent

Images