A New Variational Model for Segmentation and Denoising of Images with Multiplicative Noise I,II
Iulia Posirca, Yunmei Chen, Celia Z. Barcelos · 2015
In this paper, we develop a variational model for simultaneous multiphase segmentation and denoising of images corrupted with multiplicative noise. The presented model uses soft segmentation, which allows each pixel to belong to each image pattern with some probability being more exible than the classical hard segmentation. The denoising is performed by minimizing a variable exponential growth functional, which is a combination between TV-based and isotropic smoothing for better feature preserving. The model development and computational implementation are explored in detail, and experimental results on real and synthetic images are presented.