PROJECT TITLE :

Fingerprint Compression Based on Sparse Representation - 2014

ABSTRACT:

A replacement fingerprint compression algorithm primarily based on sparse illustration is introduced. Obtaining an overcomplete dictionary from a set of fingerprint patches permits us to represent them as a sparse linear combination of dictionary atoms. In the algorithm, we tend to initial construct a dictionary for predefined fingerprint image patches. For a new given fingerprint pictures, represent its patches in keeping with the dictionary by computing l0-minimization and then quantize and encode the illustration. In this paper, we tend to think about the impact of various factors on compression results. Three groups of fingerprint pictures are tested. The experiments demonstrate that our algorithm is economical compared with many competing compression techniques (JPEG, JPEG 200zero, and WSQ), particularly at high compression ratios. The experiments also illustrate that the proposed algorithm is robust to extract trivialities.