Automatic Fingerprint Identification Algorithm And System Research Based On Details Of The Feature Points

With the rapid development of computer science and network technique, the accuracy, practice and security of the identification technology are highly required. Biometrics based on personal identification methods has received extensive attention. Due to the uniqueness and reliability of fingerprints and the high accuracy and convenience of the technique, the automatic identification has become an attractive alternative to the traditional identification methods, and has been widely used in information safety, electronic commerce, smart phone, and criminal identification etc. In recent years, many researchers have done much work on fingerprint identification, and automatic fingerprint identification technique has been a research focus in the area of science, however many difficulties still exist.This paper analyzes the research achievements at domestic and abroad in recent years, based on the current theoretic framework, a complete algorithm of fingerprint recognition has been accomplished on Visual C++. Attentions have been given to the problems of image enhancement, image thinning, feature matching, etc for further improvements. The main contributions of my work reported in this paper are as follows:(1) For the image enhancement problems of fracture finger, an intelligent convergence enhancement algorithm of Gabor wavelet is proposed, using the properties of Gabor wavelet's direction and frequency. In the direction of fingerprint ridge of the direction field, the Gabor wavelet can make a compensation for the fracture ridge. In the vertical direction of the fingerprint ridge, Gabor wavelet fits the characteristics of fingerprint ridge and white exactly, oscillation enhancement is done to the finger ridge to achieve the purpose of fingerprint image enhancement.(2) This paper makes some improvements on OPTA(One-Pass Thinning Algorithm) thinning algorithm. The algorithm use a unified 4 ? 4 template as improved OPTA thinning structure template, improving the situation that the eliminate template and retention template is inconsistencies in former OPTA thinning algorithm. By establishing eight eliminate templates and six retention templates, the algorithm solves the problem that bifurcation is not complete and a lot of glitches generate in the ridge. Finally the experiment results show that a good result is achieved.(3)The paper has made some improvements on the fingerprint feature matching. A matching algorithm based on the combination of correcting spinal cord and polar coordinates is adopted, and the variable size limit box is introduced, therefore the algorithm is more suitable to deal with non-linear distortions. The robustness of the automatic fingerprint identification system is enhanced and the fingerprint matching speed and recognition rates are improved.