The Research And Applying Of Identification System Base On The Improved Thinning Algorithm For Fingerprint Images

Fingerprint recognition is a significant technology in biometrics. With the improvement of computer, automated fingerprint identification system be used widely in public security, finance etc. Even so, fingerprint recognition can't meet the demand for a lot of practical application yet in identifying the rate of accuracy and identifying the speed, so the performance of improving fingerprint recognition further has a very important meaning no matter in theory or in using. We not only current situation of study on recognition technology of fingerprint analyzed, do a large amount of research in design and implementation of the recognition system interconnected system of fingerprint, main contexts are as following:(1) In the preprocessing, a new thinning algorithm for fingerprint image based on the disadvantages of quick thinning algorithm and the improved OPTA thinning algorithm is proposed. The experimental results prove this processed fingerprint images of algorithm is thinning and complete, the line of skeleton is close to the core line and smoothly. So, this algorithm is a kind of comparatively ideal thinning method of fingerprint.(2) The algorithm for minutiae extraction is based on tracing the gray-level ridge, express the type of current point and the states of pixels with 8-neighborhood codes by 8-neighbour coding, then judge through checking the table method what needs to trace leaves some. Extracting minutiae and differentiating fake minutiae from the tracing result, filtering the fake minutia in the end. The experimental results show it has improved the speed of extracting minutiae, shorten time of processing images, so that can meet demands of practical application basically.(3) A minutia matching method is based on points mode matching. Revising the images according to some reference coefficient, then the data of minutia imported and template are changed into forms of radial coordinates, at last, matching data from the radial coordinates for the two and comparing the relevant degree of them by then matching string algorithm. The algorithm was proved that can improved fingerprint recognition rate.Realizing each of algorithm above with VC++ in Windows system, then applying it for the identification module of attendance of system. The experimental result had shown the high performance of fingerprint recognition system, such as processing speed fast, stability, etc. it have indicated the research of this thesis has certain practical value.