Research On Deep Learning Based On Fuzzy Theory In Face Recognition

With the development of computer technology and the increasing degree of intelligence,biometric technology is also developing at a high speed.Today,biometric technologies such as fingerprint unlocking,facial recognition,and voice recognition are widely used in various fields of people's work and life.Compared with other recognition technologies,the process of face recognition is more intuitive and has no contact with people,so it is widely used in security monitoring,smart office and other fields.It is a kind of biometrics technology that extracts relevant information from face images and uses it to identify others.Face recognition is an area of rapid development and more applications in artificial intelligence technology.At present,face recognition at home and abroad has been widely used in various fields,and has achieved relatively considerable practical results.With the increasingly common application of face recognition in real life,people are more devoted to the research of related technologies,and deep learning and fuzzy theory have become the focus of research.In this paper,the application of two theories of fuzzy theory and deep convolutional neural network in face recognition will be researched,and a fuzzy deep network model will be obtained and verified.As the amount of data in the available training set continues to increase and the accuracy continues to increase,the recognition rate is greatly improved.In today's big data era,using massive data as a basis,give full play to the powerful learning capabilities of deep neural networks,making the recognition results more accurate.This article takes the application of fuzzy deep neural network model in face recognition as the main point of research,mainly including the following:1)This article introduces the theoretical knowledge of CNN(Convolutional Neural Network)in the deep learning artificial neural network,introduces the model of the artificial neural network,and analyzes the feedforward network and the backward propagation process of the network.Explains the advantages of convolutional neural networks in image recognition,and introduces its convolutional layer,pooling layer and connection layer.2)The basic principles of fuzzy theory are studied,and the application of face recognition is studied.After analyzing the development and current status of current face recognition application technologies,it is proposed to combine fuzzy theory with deep neural networks to create a theoretical model of fuzzy deep learning;3)The combination of fuzzy theory and deep convolutional neural network is realized,and the model is verified by actual measurement with the face classic database,and the feasibility and superiority of the fuzzy deep learning model is demonstrated.