A Novel Approach For Segmentation of Good, Normal and Bad Quality Fingerprint Image

摘要

A novel approach for segmentation of good, normal and bad quality fingerprint image is proposed in this paper. Which is different from traditional methods that usually use certain features to segment diversified image. This work is divided in three steps: firstly, we enhance the quality of image to remove great many false traces effectively and got a uniform background. Secondly, we have extracted the background of ridge area and finally; we segment the foreground from background. Our results shows that the proposed technique successfully segments the good, normal and bad quality fingerprint images and also enhance the quality of image. We have evaluated the performance of the segmented image through objective and subjective quality measures like mean, standard deviation and Fourier spectrum. We have tested this technique for 3200 images taken from database FVC2002. Our technique yields better results as compared to traditional segmentation techniques. The results that we have achieved are more useful in automatic fingerprint recognition system