ROI Extraction Based on Rough Set

Region of interest (ROI) extraction plays an important part in image processing and analysis. Based on rough set theory, a ROI extraction algorithm was presented in the paper. Firstly, a rough ROI was determined based on the prior knowledge. Then, combining with the low-level features such as intensity, edge, location of the marked ROI, an information table reflecting the relation of classification was constructed and the basic regions were built based on its indiscernibility. Finally, the approximate region to the original rough ROI was considered as the final ROI. The experiments show that the new algorithm has higher accuracy and lower time complexity than the traditional methods.