Gray-scale Fingerprint Image Enhancement

Fingerprint is unique identifiers of individuals and has been used for a long time.Govemment agencies (such as the police) and many kinds of departments needing high security have to collect increasing amount of fingerprints for transmission, storage, and automated analysis.This paper arises from the practical project cooperated with the company in shanghai. What we do are fingerprint preprocess and enhancement.Get ride of the background and unrecoverable region can not only reduce the time of preprocessing, but also improve the reliability of feature extraction considerably. We compare kinds of usual fingerprint image segmentation methods, and at last achieve improved algorithm based on orientation, the outcome is not bad.Orientation is peculiar in fingerprint image, and it's fatal to the outcome of enhancement. We introduce a continuous orientation algorithm in this paper, more precise and more smooth. We optimize the program, and it's ececute efficiency is advanced. And its results are what we need in this project.We do some research on enhancement filter enhancement algorithm based on orientation. We find and solve the program that orientation templates overflow when they circumgyrate. We redesign the orientation filter accroading to this project. We study Gabor filter for comparing it with oriention filter.We get some fingerprint images that their background is very heavy, and it cannot reach the needs only by image segmentation. according to this new accident we advance a new thought getting ride of the background and noisy region in thinned fingerprint image by chain code track to solve new difficulties.