The Research Of Fingerprint Image Pos-processing Algorithm

Because of uniqueness and invariability, the fingerprint identification is becoming one of the most popular personal authentication technologies, and many fingerprint products have been offered nowadays. However, most of the key methods wouldn't be publicized, while better performance is required. So more reliable and accurate automated fingerprint identification systems (AFIS) have been of great interest to researchers.Based on plenty of papers and technical reports on fingerprint, this dissertation provides some works on the theories on AFIS. The main points are (1) A thinning algorithm of fingerprint image, (2) the minutiae extraction and false feature elimination, (3) the core detection.A thinning algorithm for fingerprint image based on OPTA thinning method is proposed. The current point is expressed the states of pixels with 8-neighborhood codes by 8-neighbour coding, then judge through checking the table method what needs to be remained. The experimental results prove this processed fingerprint image 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.The minutiae extraction is based on tracing the gray-level ridge, judging the state of the current point the ending point and bifurcate point are present respectively. For different structures of false features, a few improved methods and some new ways are applied. It is proved that tests can get exact and true features based on the elimination algorithm of false features.It is very important to detect core accurately and reliably for classification and matching of fingerprints. In this paper, a method for core detection in fingerprint images is presented to improve accuracy of the position and reliability of the core. Firstly, the approximate position ofcore is detected based on block images. Then, the inner curve properties of the ridges can determine the accurate position in thinned fingerprint image.Finally, experimental results show that these methods perform well and it is robust to poor quality images. This study systemically investigates theory and methods of automatic fingerprint identification, bring a new idea in acquiring minutiae extraction with accuracy. This study will be helpful for the people who relate with AFIS.