Object boundary detection using Rough Set Theory

A Rough Set Theory based closed form object boundary detection method has been suggested in this paper. Most of the edge detection methods fail in getting closed boundary of objects of any shape present in the image. Active contour based methods are available to get such object boundaries. The Multiphase Chan-Vese Active Contour Method is one of the most popular of such techniques. However, it is constrained with number of objects present in the image. The granular processing using Rough Set method overcomes this constraint and provides a closed curve around the boundary of the objects. This information can further be utilized in selection of similar patches for various image processing problems such as Image Denoising, Image Super-resolution, Image Segmentation etc. The proposed boundary detection method has been tested in presence of noise also. The experimental results have shown on synthetic image as well as on MRI of human brain. The performance of proposed method is found to be encouraging.