Research On Face Recognition Based On Improved Convolution Neural Network And Support Vector Machine

With the development of artificial intelligence,face recognition has received more and more attention.The research of face recognition technology has become a hot research topic.Face recognition mainly includes feature extraction and feature classification.This paper proposes improved algorithms from these two aspects respectively.(1)This paper presents a face recognition algorithm based on an improved convolution neural network and support vector machine.The convolution neural network can automatically extract the image features,but the recognition rate needs to be improved.Support vector machines have good resolution for nonlinear features.In this paper,the convolution neural network and support vector machine are combined,and the Fisher linear measure function is added to the convolution neural network to increase the distance between classes and reduce the distance within the class.The experimental results show that the proposed algorithm improves effectively.(2)This paper presents a hybrid kernel support vector machine(SVM)face recognition algorithm based on particle swarm optimization.The algorithm mixes the global kernel function and the local kernel function into a new kernel function in proportion to ensure the globalization of the global kernel function and the strong learning ability of the local kernel function simultaneously.At the same time,the particle swarm algorithm is used to quickly find the optimal solution.The characteristics of optimization support vector machine nuclear parameters.Experiments show that the algorithm also has a good recognition effect.(3)In this paper,the support vector machine of mixed kernel is applied to the convolution neural network as classifier.Then a hybrid kernel support vectormachine(SVM)model with improved convolution neural network and particle swarm optimization is proposed.