Research On Queued System Of Intelligent Push Service Based On Face Recognition

Most of the traditional Web push service is based on analysis the system's login account,history or subscription records to determine the content of the push.Since the introduction of Web2.0 technology,based on that technology of intelligent,personalized push service and application of the field of research hotspot.According to the actual needs of the personalized service of the service hall,this paper discusses the combination of face recognition technology and service push technology to realize personalized and intelligent service push system based on face recognition.In the actual push service system,a user usually only has one picture in the sample library,so this paper will focus on single-sample face feature extraction algorithm and recognition technology.This paper from the merits and demerits two aspects analysis the single-sample face recognition algorithm based on principal component analysis(PCA),and came up with a new algorithm,that first of all,the global feature is extracted by SPCA algorithm,the local features are extracted by M(2D)2 PCA+PCA algorithm,then through the maximum membership degree were obtained based on the global characteristics and local feature recognition results,finally,the fuzzy synthesis method is used to weight the fusion at the decision layer.And in the ORL face database and Yale face database on the algorithm validation.The experimental results show that the algorithm has a good recognition performance,the recognition rate of ORL face database is 89.94%,and the recognition rate of Yale face database is 90.94%.And in the expression,attitude,light and other changes et aspects have a good robustness.In order to provide more accurate personalized service push,based on user personal data analysis personalized service technology is discussed in depth,and the user collaborative filtering algorithm(UBCF)model is established.Finally,service hall's take the machine system for example,developed based on face recognition intelligent push service to take the prototype system.That system consists of four parts:face recognition,account management,information acquisition and service push.It realizes the personalized and intelligent purpose of service push system.