Research On The Embedded Automated Fingerprint Identification System

Fingerprint verification is an important biometric technique for personal identi- fication. And it is also the maturest branch of biometric technique. As a rapidly evol- ving technology, fingerprint verification has been widely used in the field of criminal identification, and obviously it has the potential to be widely adopted in a very broad range of civilian applications such as banking security. The combination of fingerprint verification and embedded systems would make this technique be suitable for many kinds of occasions. This paper did some elementary research on embedded automated fingerprint identification system, it mainly includes the following several aspects:Firstly, analyse the study actuality and development direction of this task, and explain the necessity of this task. Thinking of the high performance and low cost of ARM-chip, in this paper we construct an AFIS based on the processor Intel XScale PXA255 and the embedded operating system Windows CE. net.Secondly, this paper gives some introduction of Windows CE. Net. We design the hardware of the system, select the chips , confirm their working mode and draw the corresponding datasheet. As the software aspect, we chose the classical steps of image enhancement, binary, thin, and minutiae extraction. The paper analyses the idea, advantages and disadvantages of available algorithms of the steps, selected the best one of each step, constructs a faster, new algorithm which can thin the image of fingerprint.