Common Spatial Pattern for Human Identification Based on Finger Vein Images in Radon space

Finger vein is one of the most fitting biometric for identifying individuals. Inthis paper a new method for finger vein recognition is proposed. First the veins areextracted from finger vein images by using entropy based thresholding. In fingervein images the veins are appeared as dark lines. The method extracts veins as well,but the images are noisy, that means in addition to the veins they have some shortand long lines. Then radon transformation are applied to segmented images. TheRadon transform is not sensitive to the noise in the images due to its integral nature,so in comparison with other methods is more resistant to noise. For extractingdominant features from finger vein images, common spatial patterns (CSP) isapplied to the blocks of radon transformation. Finally the data classified by usingnearest neighbor (1-NN) and multilayer perceptron (MLP) neural network. Theresearch was performed on the Peking University finger vein dataset. Experimentalresults show that 1-NN using CSP, with detecting rate 99.6753%, against MLP ismost appropriate for finger vein recognition.

کلیدواژه ها:

Finger Vein Recognition ، Local Entropy Thresholding ، Radon Transform ، Common Spatial Patterns(CSP) ، 1-Nearest Neighbour(1-NN) Classifier ، Multilayer Perceptron (MLP) Neural Network