Palm-dorsa vein image enhancement algorithm using webcam for biometric system

Abstract

In this article, an algorithm for palm-dorsa vein image enhancement is proposed. Instead of using expensive high-quality vein image capture device, a low-cost webcam was used to capture the vein image. The webcam has been modified to make it possible to capture the vein image of palm-dorsa. In this research, the algorithm to enhance the low-quality vein image, remove the noise and finally detect the palm-dorsa vein pattern has been developed. First, the vein image is applied with image normalization technique to obtain standard image. Then, the Gaussian filter is used to remove the noise in the vein image. Next, Difference of Gaussian and threshold techniques are used to segment the vein image. Median filter is used to remove and smooth the noise introduced from the previous image segmentation process. Finally, thinning technique is used to get single line palm-dorsa vein pattern. Even though the vein images from webcam are noisy and very low contrast, results show that the proposed algorithm still can get good vein image enhancement.