Research On Face Recognition Algorithms Based On Fusion Of Monogenic Binary Coding And Deep Learning

In recent decades,with the development of science and technology,people's demand for artificial intelligence has increased day by day,and biometric technology has received much attention as part of it.Among them,face recognition has attracted the attention and research of domestic and foreign scholars because of its convenience,concealment and low cost.It is applied to all aspects,but in reality there are external factors such as occlusion,posture and illumination,and the accuracy will be reduced.Face recognition is a complex binary classification problem.The article studies the face recognition algorithm from two aspects: feature extraction and classification recognition.The following work has been done:First,the face recognition algorithm based on deep learning is studied.The efficient classifier is an important part of face recognition.The Deep Belief Network(DBN)in the deep learning model is studied.The structure and principle of the DBN network are studied based on the principle and structure of the Restricted Boltzmann machine(RBM).Secondly,In order to solve the problem of ignoring its local structural features and lacking its rotation invariance learning when extracting face image features from deep learning,an efficient face recognition method based on Monogenic Binary Pattern(MBP)and deep learning is proposed.The theoretical basic knowledge of monogenic binary patterns is expounded in detail.Based on the representation of monogenic signal of face image,the monogenic amplitude encoding rule is studied.The obtained MBP feature is the current local feature.The extraction is quite satisfactory.Finally,combining fused monogenic binary coding with deep learning is applied to face recognition.First,a monogenic binary coding feature including image structure informationis extracted,and then the feature is used instead of the image original pixel as the input data of the deep belief network to learn,and the classification result is obtained.The experimental results of 99.17% show that the MBCF+DBN algorithm has a certain recognition advantage compared with the traditional face recognition algorithm.