Palm Vein Recognition based on 2D-Discrete Wavelet Transform and Linear Discrimination Analysis

Palm Vein Recognition (PVR) is a promising new biometric that has been applied successfully as a method of access control by many organizations that has even further potential in the field of forensics. The palm vein pattern has highly discriminative features that are difficult to forge because of its subcutaneous position in the palm. Despite considerable progress and a few practical issues, providing accurate palm vein readings has remained an unsolved issue in the biometrics. In this paper, we propose an improved scheme of palm vein recognition method based on the Two Dimensional Discrete Wavelet Transform (DWT) and Linear Discrimination Analysis (LDA). The palm vein image is first subjected to 2DDWT decomposition. Then the low frequency sub-bands approximate of the image is used as an input for LDA algorithm. LDA is a subspace projection method that aims to maximize between class covariance and minimize the within class covariance. It is used initially to reduce the features, and later followed by the matching procedure using cosine distance nearest neighbor. Based on our experiments, the method has produced an identification rate of 99.74% and 100% of verification rate and 0.0% of Equal Error Rate (EER) on 380 different palms from the hyperspectral PolyU database. The total images of 3800 were captured with the wavelength of 850nm and the performance of the proposed method was better compared to images extracted using the Gabor filter method.