Automated system for fingerprint image enhancement using improved segmentation and Gabor wavelets

This research revolves around the fingerprint image enhancement. It is used for automated fingerprint identification systems (AFIS) for extracting the best quality fingerprint images. Accurate feature extraction and identification is the basic theme of this enhancement. This paper is on the fingerprint image enhancement using wavelets. Wavelets are famous for their special localization property and orientation flow estimation. The proposed technique is basically comprises of three main steps: segmentation followed by image sharpening and then Gabor wavelet filtering. Segmentation distinguishes between image background and foreground which in turn reduces processing time. Our sharpening stage of enhancement algorithm sharpens the edges and features by using prewitt mask followed by Gabor wavelet in order to enhance the feature of sharpened image. Gabor filters require orientation estimation and frequency for the improvement of the fingerprint image. But our algorithm is self sufficient. We have tested our algorithm on Fingerprint Verification Competition (FVC) 2004 database. Experimental results show that our algorithm proved to be effective in enhancing the fingerprint image quality.