Minutiae extraction is one of the core techniques of automatic fingerprint identification. Routine algorithm for minutiae extraction needs to restore ridges firstly by the way of ridge tracing on thinned fingerprint images and then minutiae extraction is realized. It is trivial and time consuming to restore ridge structure. Aiming at the problem

an improved algorithm for minutiae extraction is brought out. First

the set of original minutiae is directly extracted from thinned fingerprint images. Second

various noises in fingerprint images and their properties are analyzed and distributing regulation of pseudo minutiae is generalized. Last

combining with the information of local ridge direction

special algorithms are designed with respect to various noises to delete pseudo minutiae from original minutiae set. The remainders of original minutiae are viewed as intrinsic minutiae. To verify the performance of the improved algorithm brought out in this paper

contrastive experiment was conducted with routine algorithm. Experimental results indicate that computational time is reduced effectively with the improved algorithm and the accuracy of minutiae extraction can fill the demand of application.