Adaptive Image Normalisation Based on Block Processing for Enhancement of Fingerprint Image

초록

Improved algorithms for enhancement of fingerprint images, which have the adaptive normalisation based on block processing, in the automatic fingerprint verification system, are proposed. To obtain an enhanced fingerprint image, first an input image is partitioned into sub-blocks with the size of K×L and the region of interest of the fingerprint image is acquired. The parameters for the image normalisation are then adaptively determined according to the statistics of each block. Utilising these parameters, the block image is normalised for the next process. The proposed algorithms are tested with the NIST fingerprint images and verified to have superb performance.