Design And Implementation Of An Embedded System Of Human Face Recognition

Authentication is often encountered in daily life problems. In the electronic and information technology are rapidly developing today, areas such as e-commerce, Internet banking, public safety put forward new requirement sin in the reliability of authentication, the traditional authentication methods is unable to meet these requirements. People face as a characteristic with the advantages of difficult to counterfeit, would not be lost and portable。Compared with fingerprints, retina, iris, genes and other human biometric identification systems, face image recognition system even more direct, friendly, without any psychological obstacles for users.This article is targeted at the core technology and key solution to the problem of the face image recognition authentication system based on embedded system, to develop a research, and obtained a suitable embedded system authentication program based on face image recognition. The main contributions are as follows:(1) In2001,Viola and Jone have introduced an important algorithm that called fast face detection based on Adaboost. At PC plat form, they can meet real-time detection demand. Now we want to know whether this algorithm can achieve same detection rate at embedded system.(2) Proposes a new face recognition algorithm based on modular principal component analysis(modular PCA)together with an improved measurement of the distance of sub faces. Compared with the traditional PCA method, it can obtain a higher recognition rate when the illumination varies widely.(3) On the basis of investigation about the domestic and overseas Present status of face recognition and embedded system, firstly designed the hardware system, selected the processor chip, camera transducer and related Peripheral circles, built an embedded hardware platform with S3C2440as core.(4) Transplanted the Linux kernel, developed drivers and application programs of USB, touch serene and camera, Then develop graphical user interface..Stating from the software implementation in the article, the PCA algorithm and Adaboost algorithm have been studied and realization. After that an embedded system has been built basing on the ARM board for the human detection algorithm, then put the face detection on the ARM board.