A Novel Approach for Palmprint Ridges Features Extraction

Ridges, compared with other features extracted from palmprints, contain plentiful key traits that can be used for personal verification. This paper proposes a palmprint matching method based on ridges´ orientation and frequency information. With an optical acquisition device, high-resolution palmprint images can be employed to our palmprint verification approach. In preprocessing, palm type identification has been introduced to speed up feature extraction process. The proposed filter-based algorithm uses a bank of Gabor filters to capture both local and global details in a palmprint and represent the extracted features as different point sets. A congruent triangle method has been used to the alignment of point sets. Moreover, the match method is designed according to the character of the feature points. From the experiment, it is proven that the algorithm is of high recognition accuracy, and the EER could be as low as 1.5%.