Finger Vein Image Enhancement Technique based on Gabor filter and Discrete Cosine Transform

Abstract

Biometrics is a global technique to establish the identity of a person by measuring one of their physical or behavioral characteristics such as fingerprint, signature, iris, voice and face. Compared to these biometric techniques, the finger vein technique has distinct advantages as it helps to protect privacy and anonymity in automated individual recognition. Many studies showed that the finger vein images were of a low quality because of the variation in the tissues and uneven illumination. Hence, there is a need for effective image enhancement techniques, which can improve the quality of the images. In this study, we proposed a novel technique, which enhances the image quality of the finger veins. This method includes contrast amelioration, use of Gabor filters and image fusion, which generates an image with highly connective patterns. We used three criteria to evaluate the quality of processed images, the mean of grey values, the image entropy, and the image contrast. The obtained result shows higher values when using our approach in comparison to the baseline methods considered in this work.