Research On Embedded Application Based Fingerprint Verification Technology

Because human fingerprint is permanent and unique, fingerprint verification has become one of the most widely applied techniques in biometric verification technology and the technique of identifying people through fingerprint minutiae is becoming more and more popular. However, as fingerprint verification algorithm is usually complicated and its processing demands processors of high frequency, the high cost has affected the application development and popularization of fingerprint detection and verification technology. This research project focuses on the optimization research of fingerprint verification algorithm so that the algorithm can be processed with processors of lower frequency. In addition, a fingerprint detection and processing module board of general type has been designed based on embedded application. On the whole this dissertation can be divided into five chapters.The first chapter illustrates the aim and significance of research on fingerprint verification, reviews briefly summarization of fingerprint verification technology and puts forward the content of this research project.The second chapter is a review of fingerprint detection technique and fingerprint verification algorithm. In the part about fingerprint detection technique, the emphasis is laid on optical fingerprint sensor ,semiconductor fingerprint sensor and ultrasonic fingerprint sensor; in the part about verification algorithm , the emphasis is laid on image-based fingerprint verification algorithm, Gabor function based fingerprint image enhanced algorithm, Directional Filter based fingerprint image preprocess and fingerprint minutiae matching algorithm. At the end of this chapter is an introduction to fingerprint minutiae classifications and the definition of fingerprint minutiae used in this research project.The third chapter proposes a new algorithm to extract fingerprint minutia directly from gray-level fingerprint images. Unlike traditional fingerprint minutiae extraction algorithm which is time-consuming and requires huge image filtering and thinning processing, this new algorithm involves only two steps. First a comparatively simple fingerprint image regularizing...