Research And Application Of Fingerprint Identification System

Digitized campus, based on modern advanced computer technology, network and communication technology, constructs a virtual campus. It makes campus information resources digitized, and employs scientific management for specification, integration and utilization. In the entrance of the digitized campus, there is a need for identification so as to guarantee the identity of the personnel who enter the campus. The fingerprint recognition technology is a reliable and safe biological recognition technology, which will play an important role in the campus identification.At present, the identification of the campus network is through one card solution. The problem is the card is easy to be lost and stolen. This paper explores the introduction of fingerprint identification system to gradually replace the one card solution, thus improving the safety and convenience of identification.This paper, based on digitized campus, studies how fingerprint recognition technology plays the role of identification in the campus environment. This paper also offers a block diagram of hardware circuit fingerprint recognition, and focuses on fingerprint recognition algorithm. Based on the current achievements and technology, the research is carried out in the following aspects.(1) In the hardware section, DSP chip TMS320VC5510 from TI is adopted for the processing and recognition of fingerprint image, ARM chip S3C2440 from Samsung for external control and data exchange, FPC1011F from Sweden fingerprint company for fingerprint scanner and related expansion of memory chips. This paper gives the relevant hardware configuration diagram. The basic hardware platform is provided to meet the needs of fingerprint identification.(2) It conducts research on the extraction of minutiae. Traditional singular point extraction employs algorithm of Poincare index, which may result in the problem of the mix of pseudo-singular points and the loss of true singularity. This article, by virtue of adding some restrictions, excludes spurious singularity. Minutiae extraction employs the eight neighborhood template, and minutiae mainly include endpoint and bifurcation point, which could better extract the minutiae.(3) It studies the matching algorithm. In this process, the conception of the set of minutiae from the image recognition mode is introduced, and through the design of a modified Hausdorff distance algorithm, we could identify the matching extent of fingerprints and the database so as to improve the recognition efficiency.(4) This paper integrates fingerprint identification system with Library entrance guard system, realizing the combination of fingerprint identification technology and digitized campus. In this way, the security of digital campus will be enhanced, and the life of teachers and students on campus will be facilitated. So it has a practical value.