The Design And Implementation Of Smart Devices Based On Face Recognition That Monitor Children’s Behavior Management

With the development of multimedia technology in computer information,there are a large number of TV programs,movies and video games that middle and primary school children are interested in,they have great attraction and make children easily indulge in them.However,at the same time,most of educational videos are relatively boring,which leads to children’s low interest in watching,so that they can not make full use of the massive online video education learning resources.In order to control children’s use of home electronic entertainment devices and urge children to watch useful educational learning videos,this paper proposes and implements an intelligent monitoring terminal,which can control children to use home entertainment devices abstemiously(such as TV set-top boxes,wifi,computers,etc.)and urge children to watch useful educational learning videos,and for the children who can do so,they can be rewarded to use a certain of time on electronic entertainment equipment.In order to make children have good learning and life performance at home and at school,this intelligent monitoring terminal adopts incentive mechanism according to children’s performance at home and at school,so as to increase children’s learning motivation and gradually make children develop good living habits.This intelligent device mainly uses the face recognition technology of open source based on deep learning to realize its key functions,so this dissertation introduces the principles and implementation methods of face detection,face feature extraction based on deep learning,and face matching.In view of the slow operation of face detection in embedded systems,this dissertation proposes a method to speed up the detection of face location information in images.As the open source face recognition database is not accurate enough to meet the commercial requirements,this dissertation proposes a method to improve the accuracy of face recognition.For the possible cheating situation of children(such as using photos to avoid face recognition system),this dissertation presents the implementation method of bio-assay detection.Based on the RK3288 chip of ARM,the off-line face recognition function is realized so as to avoid connecting cloud to recognize.On this embedded platform,the design and development of the hardware platform and software system of the intelligent terminal for children’s behavior management based on face recognition are realized.After full functional test,accuracy and performance test of face recognition as well as system stability test,the intelligent terminal achieves good desired results.