Research And Implementation Of Intelligent Terminal Identification Based On Dynamic Gesture

With the development of mobile Internet,smart phones have been widely used in public life.Mobile Internet has brought information sharing and the threat to users’ privacy.There are frequent breaches of user privacy data due to the theft or loss of mobile phones.Therefore,how to achieve accurate and long-term guarantee of the validity of the intelligent terminal visitors has become the focus of many scholars.But most scholars’research is based on entry type,which can’t continuously monitor the user’s legitimacy in the background after the user enters the system.In view of the above situation,this paper proposes a smart terminal identification method based on dynamic gesture,which realizes the 24-hour continuous monitoring of users as illegal users in the background.The following research contents are mainly included:1.In this paper,the method of wavelet packet decomposition and Pearson correlation coefficient is used to filter the raw data.First,adopt the method of wavelet packet decomposition of the original data are decomposed into several frequencies,each spectrum contains different information,then with Pearson correlation coefficient correlation analysis was carried out on the adjacent frequencies,dynamic hand gestures to find information and the cut-off point of high frequency noise,and then carries on the filtering de-noising.2.In this paper,the traditional k-means algorithm is optimized in the initial center number and initial center position point.The traditional K-Means algorithm is the choice of initial center number and center point completely random,in this paper,the initial center number choose to take the way of enumeration,the choice of initial center position is half a randomized way.3.This paper presents an identification method based on dynamic gestures.Grade according to the result of clustering is to generate a judgment formula,based on the new sample,also provides the new sample data recovery mechanism,the optimized model.4.In order to verify the feasibility of this method,a smart terminal identification system based on dynamic gesture is constructed.Client USES is the Android system of intelligent terminal,use this system to the experiment of the proposed approach,experimental results show that the method in the experiment FAR(False Acceptance Rate)value of 4.2%,the FRR(False Rejection Rate)value of 1.92%.It is proved that this method is feasible in practical application.