Neural-network-based facial feature extraction modeling and face recognition methods and devices

Abstract

The invention provides a neural-network-based facial feature extraction modeling method and a neural-network-based facial feature extraction modeling device. The method comprises the following steps: acquiring each block of a first picture, each block of a second picture, a first class of the first picture and a second class of the second picture from a training set of a preset application scenario; determining a global block damage cost function value corresponding to the blocks of the first picture; training a facial identity verification neural network on the training set; testing the facial identity verification neural network corresponding to the blocks of the first picture, and determining a target facial identity verification feature extraction model corresponding to the blocks of the first picture. According to the method and the device, the beneficial effect of improving face recognition accuracy when the facial identity verification feature extraction model obtained by modeling is applied to the preset application scenario for face recognition can be achieved. The invention also provides a face recognition method and a face recognition device applying the neural-network-based facial feature extraction modeling method and the neural-network-based facial feature extraction modeling device.

Description

technical field [0001] The present invention relates to the technical field of face recognition, in particular to a neural network-based face feature extraction and modeling method and device, and a face recognition method and device. Background technique [0002] Face recognition is a biometric technology for identification based on human facial feature information. Generally, a camera or camera is used to collect images or video streams containing faces, and then automatically detect and track faces in the images, and then Face recognition is performed on detected faces, which is also commonly called portrait recognition or face recognition. [0003] At present, face recognition algorithms are based on face photos and corresponding identity information, extracting features from a picture as a whole, using neural networks for model training, and finally using classifiers for face recognition. In the training of the model in the face recognition neural network, only the ide...

AI Technical Summary

Problems solved by technology

Therefore, the recognition accuracy of face recognition using this model needs to be further improved.

Images