DefinePK

Abstract

Finger vein images are prone to various noise due to imperfect capturing device used. As a result of transmission errors, pixel elements in the capturing sensors malfunction, memory location fault and analog-to-digital conversion time errors, noise could also occur. Using single filtering method cannot give a satisfactory result of the vein image. Hence, this paper proposed a multifilter algorithm that’s capable of locating the region of interest of the vein and de-noise the image from corrupted noises while the edges of the image are kept intact. The algorithm does not need to have prior knowledge of the image and the corrupted noise. The fusion of simple mask filter algorithm and mixed filter algorithm of Median and Wiener filters is used. Experimental results coded in MATLAB using SDUMLA-HMT finger vein database show that the proposed algorithm is feasible and efficient.