Fingerprint image enhancement by super resolution with early stopping

Abstract

This paper addressed the problem of multi-frame image super resolution and implementation to fingerprint image. The use of computers in fingerprints recognition is highly desirable in many applications where security is an important concern. However, as the performance of fingerprint recognition algorithm relies heavily on the quality of the input fingerprint images. it is essential to incorporate a fingerprint enhancement algorithm in the minutiae extraction module. Typical approaches use Gabor filter to raise the contrast between fingerprint ridges and valleys. But few of them try to increase the very basic resolution of fingerprint image. Therefore, we proposed a computer vision solution to this problem which involves the technique of super resolution. This approach can be incorporated into any traditional fingerprint enhancement algorithm as a pre-processing step, rendering a better overall enhancement result.