Fingerprint Image Preprocessing Algorithm

Due to their uniqueness and persistence, fingerprints are used as main basis of personal identity. Automated fingerprint identification system(AFIS), a technology of fingerprint identification using computer, is of convenience high fields efficiency, security and reliability. It has been applied in many such as financial security, data encryption, electronically business and will play a more and more important role in our life.Up till now, affected by the noise and the skin elasticity, fingerprint can not be identified efficiently. The problem how to improve the performance of Automated Fingerprint Identification will be discussed in the dissertation. In this dissertation, algorithms will be improved in the aspect:pre-processing.Fingerprint identification system include steps of fingerprint images collection, image preprocessing, feature extraction and matching. The dissertation discusses and studies a number of problems about the automatic fingerprint identification system and fingerprint identification algorithms are summarized. Then we focus on the image preprocessing and feature extraction algorithms. In this dissertation, A combinatorial algorithm based on the directional segmentation is proposed for the preprocessing of the fingerprint images, This combinatorial algorithm is composed of normalization,directional image computation, directional filtering(gabor filtering), local self-adaptive binarization, noise elimination and thinning. In gabor filter, we use eight directional filtering to process fingerprint image to compare wiener filtering, we get good process result. We compare local self-adaptive way with the fixed value in binarization, and use mathematical morphology method to thin image.Experiments show that the algorithms we mentioned can enhance the quality of fingerprint image efficiently.