Abstract: The finger vein is a unique physiological biometry to identify individuals based on the physical characteristics of vein models in humans. The technology is currently being used or developed for a wide variety of applications, including credit card authentication, automotive security, computer and network authentication, automated teller machines. The proposed technique is a fingervein recognition system based on fuzzy score level moment invariants fusion and corner detection. It is consist of two phases, enrollment phase and test phase. In the enrollment phase, several operations are implemented. Firstlly, after the preprocessing, the seven moment invariants features extracted from the input fingervein image and detect the corners locations, respectively. Second, the standard deviations of corners location of the image is calculated. Third, these features used as an input for the fuzzy rule and fuzzy membership function, respectively. Fourth, the use of a fuzzy-based method to enhance matching technique. Finally, the corner detector techniques are used for matching. The experimental results using finger-vein dataset showed that the proposed method is enhanced the accuracy of finger-vein recognition compared with previous methods.
Jane Jaleel Stephan, Entessar Karim Hanoun and Ali Makki Sagheer, 2018. Fingervein Recognition System Using Fuzzy Score Level Fusion and Corner Detection. Journal of Engineering and Applied Sciences, 13: 7990-7996.