Research Of Palmprint Recognition Based On Android System

As the emerging biometric recognition technology,palmprint recognition has developed rapidly in recent years.The problem of information security is becoming more and more serious,and many scholars have put forward the use of their own biological characteristics to encrypt it,and gets a good result.The advantages of palmprint recognition with high security and stability stands out among many biological characteristics.So this paper has carried out the research on the palmprint recognition based on Android system.1.This paper analyzes the problem of traditional palmprint recognition acquisition device and proposes an acquisition application based on Android mobile phone.For traditional non-contact positioning difficulty,this paper proposes hand shape box positioning method.The method can determine the effectiveness of the palm of your hand,overcome the hand shake with more than the depth of field image fuzzy,prevent deformation of the palm of your hand.In order to identify the image directly,the feature extraction was carried out in the center area of the split palm.The experimental results confirm that the image enhancement can improve the effect of feature extraction.2.For the identification of palmprint recognition feature extraction algorithm,this paper studies the recognition of palmprint based on the description subalgorithm.There are two main categories: SIFT algorithm and its derivation class,LBP algorithm and its improved class.In this paper,SIFT algorithm class and LBP algorithm class are compared,and the best algorithm for each class is compared,and the palmprint recognition algorithm suitable for Android phones is found.The experiment results show that the SURF algorithm can meet the real-time and accuracy,and can be used for palm print recognition on mobile phones.3.This paper studies the development and operation environment of Android system.A detailed description of the communication Intent among the four components and the four components of Android is described.According to the hardware and software requirements developed by Android system,the Android platform is built.According to the implementation requirements design system process,the software system is divided into user login stage and user registration stage.The system functions are divided into three modules: collection of registration module,feature extraction module and feature matching module.The implementation process of each module is explained in detail.Experimental tests on the complete system show that the recognition rate of palmprint identification of Android system is high and fast,which lays a good foundation for the next promotion application.The common non-contact collection has the effect of invalidation of collection image,the acquisition of palm deformation or the fuzzy acquisition of the palm,which leads to the failure of the recognition rate in the later stage.In this paper,a hand-based frame method is proposed to solve the problem of acquisition,improving the recognition rate.The design of the hand frame breaks through the palmprint recognition in the non-contact direction of the bottleneck.I hope to give some new ideas to the future researchers with better use the palmprint identification.