Research On Algorithms For Detecting Of Singularities And Fingerprint Enhancement In Fingerprint Identification System

As the rapid developing of computer technology ,Automatic Fingerprint Identification System becomes a important method by which current secureity problem which we are facing with can be solved well,but because of the in-time need in Auto Fingerprint Identification System(AFIS), currently some more advanced algorithms are influenced greatly in actual application with their complexity. So we must introduce a high-efficieny, well in-time identification algorithm to meet the current need of application.This papar's major work is how to detect singular points accurately and reliably in fingerprint images with serious noise.It also study on fingerprint enhancement.It is very import to detect singular points (core and delta) accurately and reliably for fingerprint classification and point-pattern matching of fingerprint. In this paper, on the one hand, in order to imprive quality of fingerprint and reduce computational complexity, we first process some indispensable preprocessings which include wavelet multi-resolution decomposition, image segmentation, iamge enhancement before detecting singular points,by these preprocessing, we not only can eliminate the noise during fingerprint collection, but also reduce data capacity of fingerprint greatly; on the other hand ,the paper also brings forward a new algorithm on detecting singularities, ie, in this method, the characteristic of the relative concentration of the position of singular points detected though image shift and of the corresponding relationship of the singular points detected with multilevel block sizes are used and the singular points are detected accurately and reliably. Experimental results show that the method performs well and it is robust to poor quality image.In addition, in this paper, a fast fingerprint enchancement algorithm based on gray-level image is introduced, this algorithm improves the clarification of ridge and furrow structure of fingerprint iamges self-adaptively based on local ridge frequentcy and direction. In this algorithm ,good band-pass performance of Gabor filter is employed ,ridge direction and ridge frequency of block fingerprint image is used as parameter ,Gabor filter is used to filter and enchance the image quality ,and it can eliminate the noise ,it can also increase the ridge clarification of recoverable area and make it suitable for minutiae extracting algorithm .Thisalgorithm can also distinguish corrupt areas in which the trude structure can not be recovered, it can distinguish, label and mask these areas ,which can effectively reduce the generation of spurious minutiae.