Fingerprint Recognition Algorithm

With the development of the information society, the traditional identification pattern can’t satisfyour needs. In order to meet the demand of information security, biometrics has received more and moreattention. There are many human biometric features that can be used to identify individuals, such asvoice, face, fingerprint and retinal pattern of eyes. Among all these features, fingerprints, due to theirflexibility, uniqueness and cheapness, are the most widely used biometric features.Although great progress has been made in the development of fingerprint identification technology,some difficult problems are still not solved in the fingerprint identification field. This paper focuses onthe system of the fingerprint identification algorithm. We have reading a lot of literature and tried toimprove algorithm efficiency.1、In this paper a lot research have been made on orientation field. Because the orientation fieldof a fingerprint image is so important that effects image pre-processing. We improve the traditionalorientation extraction algorithm by compressing amount of image data. Then we calculate orientationfield，so the accuracy and speed is ensured.2、In the last step of the fingerprint identification，a new matching algorithm is firstly proposed,this algorithm concludes two steps that are fast-mode and accurate-mode. Experiment proves thisalgorithm can improve the matching speed, and is guarantee recognition accuracy.3、At last a new fingerprint identification algorithm that combines point match and image matchis proposed. This algorithm depends on singular point abstraction and wavelet transform coefficient.But this algorithm actual performance is can not meet practical needs, so it required more research onit.