A new method of extracting palmprint features is presented in thisreport. Since palmprint images have many creases which are organizedlike ridges, ordinary fingerprint feature extraction algorithms areunable to extract ridges. Consequently, the goal of this research is toconstruct a new feature extraction method which can extract ridges underthese conditions. At first, the original palmprint image is divided intolocal images. The ridge candidates are extracted from each local imageof the palmprint. Finally, only one candidate is selected as the ridgein the local image by estimating the continuity of certain properties.Experiments with palmprint images indicate that the proposed method canextract ridges better than conventional methods, especially in areas inwhich both ridges and creases exist
展开▼
