Summary: | 碩士 === 國立交通大學 === 電資學院學程碩士班 === 90 === Fingerprint classification provides an important indexing mechanism in a fingerprint database. An accurate and consistent classification can greatly reduce the fingerprint matching time for a large database. In this thesis, we present a new classification method for fingerprint images. In the proposed method, we classify fingerprints into five classes: arch, left loop, right loop, whorl, and tented arch. The major steps of this method include image enhancement, direction matrix extraction, singular points extraction and classification. Finally, we use the 1900 thumb fingerprints of NIST-4 database to evaluate the performance of the proposed method. The experimental result shows that we are able to achieve a classification accuracy of 88 percent (with 10% rejection).
|