Fingerprint Distortion Removal and Enhancement by Effective Use of Contextual Filtering

Abstract

Automated personal authentication has become increasingly important in our information driven society and in this regard fingerprint based personal identification is considered to be the most effective tool. In order to ensure reliable fingerprint identification and improve fingerprint ridge structure, novel fingerprint enhancement approaches are proposed based on local adaptive contextual filtering. This dissertation presents the aims and objectives of the research along with the motivation and proposed research approaches for fingerprint enhancement mentioned below. In the first part of this research, a twofold enhancement technique is proposed that involves processing both in frequency and spatial domain. The fingerprint image is first filtered in frequency domain and then local directional filtering in spatial domain is applied to obtain enhanced fingerprint. Two major advantages of the proposed enhancement technique are; avoiding the hazards of frequency estimation errors and making spatial domain filtering quite simple by using a global 1D Gaussian smoothing filter. In order to determine the performance evaluation of the proposed enhancement, extensive evaluation of the proposed method against well-known enhancement approaches has been carried out on publicly available standard databases. Experimental results demonstrate that proposed enhancement performs better as compared to other well-known enhancement techniques. In the second part of this work, a fingerprint image with non-uniform ridge frequencies is considered as a 2-D dynamic signal. A non-uniform stress on the sensor’s sensing area applied during fingerprint acquisition may result in a nonlinear distortion that disturbs the local frequency of ridges, adversely affecting the matching performance. This study presents a new approach based on short time Fourier transform (STFT) analysis and local adaptive contextual filtering for frequency distortion removal and enhancement. In the proposed approach, the whole image is divided into sub-images and local dominant frequency band and orientation are estimated. Gaussian Directional band pass filtering is then adaptively applied in frequency domain. These filtered sub-images are then combined using our proposed technique to obtain the enhanced fingerprint image of high ridge quality and uniform inter-ridge distance. Experimental results show efficacy of the proposed enhancement technique.