Design of Facial Recognition System for Smart Home Mobile Based on Deep Learning
In order to improve the detection accuracy of face recognition system in smart home mobile,a multi-task deep learn-ing algorithm based on convolutional neural network is proposed in this paper.Through convolutional neural network,multi-level features in face images are extracted,and the local differences of faces are captured and distinguished.The face images are collected and preprocessed,and the MTCNN model is constructed to obtain the frame and the coordinates of key points.The joint loss function is used to optimize the multi-task,which can complement and enhance the information among the tasks and improve the recognition accuracy of the single task.Then,the generalization performance of the model is effectively enhanced by combining the multi-task learning framework.The experimental results show that the recall rate is over 98.3％and the recognition accuracy is up to 95.6％,which is significantly better than the existing system.
deep learningsmart homemobile terminalfacial recognition
NETL
NSTL
万方数据
