Images can be broadly classified into two types: isotropic and anisotropic. Isotropic images contain largely rounded\nobjects while anisotropics are made of flow-like structures. Regardless of the types, the acquisition process introduces\nnoise. A standard approach is to use diffusion for image smoothing. Based on the category, either isotropic or\nanisotropic diffusion can be used. Fundamentally, diffusion process is an iterated one, starting with a poor quality\nimage, and converging to a completely blurred mean-value image, with no significant structure left. Though the\nprocess starts by doing a desirable job of cleaning noise and filling gaps, called under-smoothing, it quickly passes\ninto an over-smoothing phase where it starts destroying the important structure. One relevant concern is to find the\nboundary between the under-smoothing and over-smoothing regions. The spatial entropy change is found to be one\nsuch measure that may be helpful in providing important clues to describe that boundary, and thus provides a\nreasonable stopping rule for isotropic as well as anisotropic diffusion. Numerical experiments with real fingerprint data\nconfirm the role of entropy-change in identification of a reasonable stopping point where most of the noise is\ndiminished and blurring is just started. The proposed criterion is directly related to the blurring phenomena that is an\nincreasing function of diffusion process. The proposed scheme is evaluated with the help of synthetic as well as the\nreal images and compared with other state-of-the-art schemes using a qualitative measure. Diffusions of some\nchallenging low-quality images from FVC2004 are also analyzed to provide a reasonable stopping rule using the\nproposed stopping rule.
Loading....