A method for hand vein recognition based on Curvelet Transform phase feature

A new approach to palm-dorsal vein recognition based on the second generation Curvelet Transform is presented in this paper. After palm-dorsal image preprocessing and ROI (Region Of Interest) extraction, we use UDCT (Uniform Discrete Curvelet Transform) of the Curvelet Transform on ROI, and encode the Curvelet coefficients phase variance, and evaluate the Chi-square distance of coding histogram for vein recognition. Experiments are carried on our low-resolution palm-dorsal vein image database including 400 images. Experimental results show that UDCT phase feature has higher recognition rate (EER is 2.17%) compared with FDCT phase feature, PDFB phase feature, Gabor phase feature [17], and Triangulation and Knuckle Shape [9]. UDCT phase feature consumes less time than FDCT phase feature, Gabor phase feature, and Triangulation and Knuckle Shape.