Evaluation of the Fingerprint Images Captured from Optical Fingerprint Scanner

摘要

In this paper, we propose the way to evaluate the quality of fingerprint images captured from optical fingerprint scanner. Fingerprint image-quality can be digitized by using the "measure" we proposed. We simulate using the database consists of 1425 fingerprint images captured from 57 people in February, which contains a lot of faded images. In the simulation using all our database, recognition rate is 95.6 [%] while type II error is 0.01[%]. Recognition rate is improved to 98.1[%], with rejecting 3.7[%] faded images evaluated by our measure from the database. Recognition rate is improved to 99.6[%], rejecting 14.2[%] faded images. And we investigate the way to keep a balance between verification accuracy and user-friendliness of fingerprint verification devices with applying the measure.