Fingerprint image enhancement based on skin profile approximation

The performance of the fingerprint identification and verification systems relies heavily on the quality of the input fingerprint images. In this paper, we propose an effective image enhancement algorithm especially suitable for low quality fingerprint images, which can improve the clarity and continuity of the ridge structures. Unclean sensor plates, non-uniform and inconsistent contacts are among the major causes for poor samples and feature extraction artifacts during image processing. The proposed algorithm estimates the finger elasticity by approximating the uneven finger skin due to poor skin condition or imperfect acquisitions. Then the fingerprint image is normalized with respect to the approximation. Experimental results show that the enhanced image quality by using the proposed normalization algorithm is effective and much better than other existing methods for improving the minutiae detection.