Robot Human Body Comprehensive Feature Recognition Based On GPU Cluster

Personal identity authentication is an important research topic in the field of computer vision.In recent years,artificial intelligence robots,especially intelligent service robots,have begun to be applied in people's daily life,including: security patrol,government services,family services,Banking services and medical assistance.In order to provide robots with more stable intelligent service performance,robots need to have the ability to accurately and quickly identify people.In this paper,the machine learning method is used to study face recognition and gait recognition and its integrated human body comprehensive feature method.The main contents are as follows:(1)Aiming at the fast and accurate face detection problem,a cascaded neural network based on improved MTCNN is proposed.By simplifying the model structure,the detection speed can be improved while ensuring the detection accuracy.Aiming at the problem of face feature extraction,the feature extraction based on traditional algorithm and the feature extraction method based on deep learning network are introduced.Aiming at the face feature matching problem,the SVM classification algorithm and the classification method based on cosine distance calculation are introduced.In order to improve the matching rate,a matching optimization method based on age and gender identification is proposed.(2)Aiming at the gait recognition problem,the recognition method based on LBP and HOG feature layered fusion is introduced.The local binary model and gradient histogram are used to extract the features of several layers of the target image,and the features of each layer are merged and sequenced.The joint becomes the final feature for matching recognition.(3)Aiming at the problem of gait and face fusion recognition,the gait and face fusion recognition methods based on feature layer and decision layer are introduced.Firstly,the image is preprocessed,then the gait and face feature extraction is performed,and the gait and face fusion are used for identity recognition.The related algorithms in this paper have passed the system test and verification.The results show that the proposed algorithm achieves the expected requirements in the character authentication effect and lays a foundation for the system application of intelligent service robot.