Palmprint features extraction using wavelet transform

The ridge orientation and the ridge spatial frequency in different regions represent the intrinsic nature of the palmprint image. To extract the palmprint features, a new algorithm based on wavelet transform of the original gray scale image is proposed. The features are extracted directly from the gray scale palmprint image without preprocessing (i.e. image enhancement, filtering, region segmentation, binarization, ridge thinning, etc.); hence, the proposed algorithm achieves lower computational load than the traditional minutiae-based algorithms. The efficiency of the new algorithm is validated with the high recognition rates achieved by testing in a small palmprint database using the K-nearest neighbor (K-NN) classifier.