A Novel Method for Fingerprint Image Segmentation Based on Adaptative Gabor Filters

The human fingerprint offers a reliable authentication mechanism that allows identify users based on natural features. One of the issues found in the feature extraction process for automatic fingerprint identification is the delimitation of the region of interest in a fingerprint image. This paper presents a novel procedure to extract the region of interest of a fingerprint based on Gabor filters. Tests using datasets provided by the Fingerprint Verification Competition were performed in order to evaluate the proposed methods performance (execution time and error rates) and to compared it with related work in the literature. The results showed lower error rates but higher execution time for the proposed method, when compare to competing methods.