Abstract
Automatic fingerprint recognition has received considerable attention over the past decades. Progress has been made on models of the structures of fingerprints, techniques for acquisition of prints, and the developments of commercial automated fingerprint identification systems. Despite these advances, there remain considerable opportunities for improvements. The speed of retrieval and the ability to recognize partial or distorted prints are prominent among those areas that require improvements. This paper discusses a new technique to remove scars from impaired fingerprint images. This is achieved by first improving directional images of finger prints and then followed by reconstructing the finger print images using Inverse Fast Fourier Transform. The technique was tested on 500 fingerprint images of 5 different classes namely arch, left loop, right loop, tented and whorl. The experimental results have shown that the reconstructed images have achieved almost 100% free of scars and have a better and clear ridge and furrow structures.