FINGERPRINT FREQUENCY ESTIMATION

Frequency information is very important in the enhancement of the fingerprint image while it is very difficult to be estimated accurately in some areas. In this paper, a new and robust frequency estimation algorithm is proposed which is completely independent of the local orientation information and ridge structure. As a narrowband texture image, there are different output responses through different scale Gabor filters and the maximum response will correspond to the right frequency. The experimental results show our algorithm is accurate and it provides a novel approach of fingerprint frequency estimation.