Polyphase adaptive filter banks for fingerprint image compression

Subband decomposition is widely used in signal processing applications including image and speech compression. In this paper, we present Perfect Reconstruction (PR) polyphase filter bank structures in which the filters adapt to the changing input conditions. This leads to higher compression results for images containing sharp edges such as fingerprint images. The fingerprint image compression is an important problem due to the high amount of fingerprint images in databases [I]. For example, the FBI database contains 30 million sets of fingerprints. We experimentally observed that our method is successful for binary and gray-valued fingerprint images.