Research On Face Recognition Based On Improved LBP

Face recognition technology is one of the important research directions of biological recognition technology,and making computers has the abilities of identification and identification is the ultimate goal of the technolo gy.Many departments institutions,such as medical treatment,monitoring,security and financial,requirements of information security are increasing,and these requirements make people focus on the facial recognition technology,which was widely used.So this technology has become the daily life as one of the most widely used biometric identification technology.Local binary pattern(LBP)is a method which described image texture by comparing the pixel values,at the beginning,it is an assistant method used to measure the local contrast of an image.Algorithm constantly evolves and is used in the facial features description and recognition,and studies have shown that the implementation of the effect is very good.However,the LBP operator itself still has some shortcomings,and to solve some of the problems in face recognition further studies are still needed,to solve these problems,this article completed the research work mainly in the following two aspects:1.In the process of feature extraction,the Local Binary Pattern(Local Binary Pattern,LBP)operator processing principle and human face feature extraction process was studied,and then combined with the prior knowledge of the distribution of facial features,this paper presents a new partition method based on LBP algorithm,in order to obtain more effective characteristic values,and improve the recognition rate of face recognition algorithm.2.In the process of face recognition,the training sample and comparing similarity of unknown sample is identified as a result.The identification results were recorded,and after the comparing recognition rate of block,this paper puts forward the screen process to get the characteristics which have universality,and the characteristics format a new characteristics profile.Updating the profiles,the recognition rate of face recognition is improved.