An optimal method for identification of finger vein using supervised learning

The identification of finger vein is very advanced biometric technique and also behove an important discipline in biometric area, accommodating growing attention in last years. In the same manner with the help of vein design allows high security of personal identification. In this research, we have applied PCA (principal component analysis) for extracting vein patterns and SVM (support vector machine) for classification. The method is purely depends on the different pattern qualities of the vein. The best optimal design has been used in developing the efficient framework. We are using the three publicly available datasets in the research i.e. THU-FVFDT2, SDUMLA-HMT, and FV-USM. Our results clearly indicate the superiority of our proposed framework.