Palmprint Authentication System Using Wavelet based Pseudo Zernike Moments Features
Ying Han Pang, Andrew Beng Jin Teoh, Beng Jin, David Chek Ling Ngo, Chek Ling · 2005
In this paper, a novel method of image based palmprint matching based on features extracted from wavelet-based pseudo Zernike moments feature descriptor is proposed. Pseudo Zernike moments have additional properties of being more robust to image noise, possessing geometrical invariants property, having a nearly zero value of redundancy measure in a moment set due to its orthogonality property and having a superior image representative capability; while wavelet analysis affords major advantages of performing local analysis (describe image local characteristic) due to its excellent localization property and decomposing image to ease image information interpretation. Therefore, the hybrid wavelet transform and pseudo Zernike moments is able to comprise salient features from both imaging apparatus and achieve better verification rate. Comparison analysis shows that the hybrid wavelet transform and pseudo Zernike moments is able to achieve superior performance than the other wellknown moments.