Research On Pre-processing Algorithm Of Fingerprint Image

The identification authentication is ubiquitous now in the society. Because of the flaw of the traditional status authentication method, the encryption intensity and the application method all more and more cannot be satisfied with the request of the modern society. Along with the computer technology development, the development of the biological recognition technology gradually enters to the practical stage. In field of fingerprint recognition, so many researches and experiments are made by scholars home and abroad that plenty of algorithms very useful in this field are proposed and a set of stable process are formed.The primary topic of this thesis is the pre-processing algorithm of fingerprint. First of all, introduced the fingerprint history, the characteristic, the classification and so on, then with fingerprint image characteristic, each stage of the pretreatment process is elaborated comprehensively, namely fingerprint image intensification, binaryzation, refinement and refinement post-processing and the other classical algorithms, and the algorithms are carries on the analysis through the experiment. However finally the still existed problems are proposed in the fingerprint image.(1) Fingerprint image enhancement:The fingerprint image histogram, histogram equalization and gray level change approach has reached the fingerprint image enhancement purposes.(2) Fingerprint image filtering denoising:how many image filtering denoising method to select a most suitable method of fingerprint image.(3) Binary image:the fingerprint image from the gray image into binary image, which is an important part, it will directly affect the follow-up treatment.(4) Fingerprint image refinement:refinement main role is to remove unnecessary information ridge thickness, so that the fingerprint image data and continuous structural change is simple, easy to extract details from the characteristics of the fingerprint image, thereby enhancing the fingerprint image Processing speed and efficiency.(5) Thinned image post-processing:how to eliminate false details, reduce errors matching opportunities.Finally, also proposed for automatic fingerprint extraction current, there are still problems, these problems will be corrected in future research.Finally, also proposed for automatic fingerprint extraction current, there are still problems, these problems will be corrected in future research.(1) Sometimes the filter denoised image is more blurred edges while filling, is not conducive to the conduct of follow-up needs to be further explored to retain very good denoising images while the high frequency component of the filter.(2) Threshold selection varies, which greatly limits the generality of the algorithm.(3) Fingerprint ridge after refinement is not consistent, but also requires further work.