Research On Face Recognition Algorithm Based On Improved Neural Network

Face recognition technology is a very important issue in the field of artificial intelligence,and it is also one of the hot topics in current research.This paper focuses on the research of face recognition algorithm based on neural network under the influence of external factors.A face recognition method with strong stickiness and high recognition efficiency.The main research contents of the thesis are as follows:(1)We used principal component analysis to initialize the convolution kernel in the convolutional neural network,the complexity of convolution calculation is reduced,and the running time is reduced.Secondly,the improved learning rate is used to optimize the convolution layer weight update.The experimental results show that the optimized neural network recognition rate is significantly improved.(2)The face data set is preprocessed by using the method of scale normalization and gray normalization.The experimental results show that after normalizing the image data,and it obviously reduces the influence of illumination on face recognition,which is more conducive to extracting the features of face.(3)We proposed a convolutional neural network face recognition method based on Sobel operator preprocessing.Then the improved network is compared with PCA,BP neural network and traditional CNN face recognition method.The experimental results show that the improved network recognition rate is significantly improved and the recognition efficiency is significantly accelerated.