Abstract
Fingerprint identification systems are one of the most well-known and publicized biometrics because of the inherent ease in acquisition, the numerous sources (ten fingers) available for collection, and their established use by law enforcement and immigration. These systems rely on the unique biological characteristics of individuals to accurately verify their identities. To get reliable and accurate verification results, these systems need high quality images. The quality of the fingerprint image is obtained by using noise-free images during the pre-processing and filtering stages. In this paper, we proposed an integrated smoothing method (ISM) for fingerprint image recognition enhancement based on a linear combination of three different filtering techniques named median filter (MF), Wiener filter (WF) and anisotropic diffusion filter (ADF). This combination is made by using two coefficient parameters (alpha,beta) with different values to enhance the quality of images and remove the unwanted distortion or noise that affect a fingerprint recognition system. The ISM is applied in the pre-processing stage to get a noise-free fingerprint image with high accuracy factor. We used the benchmarking FVC2004 and FVC2006 databases to test our method and the Wilcoxon signed-rank test (W) and the peak signal-to-noise ratio (PSNR) for results evaluation. The experimental results indicate that the proposed ISM improves the performance of the fingerprint identification significantly.