Research And Realization Of Fingerprint Identification Algorithm

Recently, the Biometric technology has been universally noticed, and it is considered that this technology has a great future. Now, Fingerprint Identification is the most developed Biometric technology extensively used in many fields, which has a bright future. Due to their uniqueness and persistence, fingerprints are used as main basis of personal identity. Automated fingerprint identification system, a technology of fingerprint identification using computer, is of convenience, high efficiency, security and reliability. It has been applied in many fields such as financial security, data encryption, electronical business and it will play a more and more important role in our life. Based on latest extensive discourses and technology journals in this field, the dissertation is trying to make studies on algorithms of AFIS. Main works are as follow:The research of fingerprint's collecting system. The system adopted DSP processor as a fingerprint processing chip, and adopted a silicon crystal sensor as fingerprint collection, CPLD as global logic control, and designs a host-slaver USB interface, not only realize the transmission between DSP and PC, but keep the data on U dish divorced from PC. Each part works in phase, constituting an integrity of hardware system.The research of fingerprint image's pre-processing. Its purpose is removing noise of fingerprint image and improving fingerprint definition of fingerprint image, getting the structure of image to a one gels width image, which will be beneficial to the follow feature extraction. It includes image normalizing, orientation estiation, frequence estimation, Gabor filter, binary, thining and removing noise after thining.The research of fingerprint image's minutaes extraction and matching. Minutiaes extracting mainly got fingerprint minutaes and position. Firstly, this dissertation uses 8-neighborhood cross-calculation algorithm to extract all minutiaes from the pre-processed image, and we validate these minutiae and eliminate pseudo ones through ridge tracking method. Fingerprint image matching is judging whether these two fingerprint images are captured from the same finger. We have developed anfeature point-matching algorithm. The matching of point is transformed into matching of vectors, which are composed of two feature points.The dissertation researched the algorithms of pre-processing and minutaes extraction etc. The experiments showed that the algorithms can be effectively enhance fingerprint images, and can accurately extract fingerprint image's minutaes.