A new approach for image segmentation using improved k-means and ROI saliency map

摘要

The very crucial part of image processing system is Image Segmentation (IS). In the IS process, the accuracy is dependent on segmented image. In this paper, a new method for image segmentation is shown using improved K-means and region of interest (ROI) saliency map. The proposed system is classified the brain MRI and breast MRI pictures. In this process, detect the portion of tumor from MRI image. By the use of ROI saliency map algorithm, detect only circle section of the picture on the basis of connected component. The experimental dataset contains brain MRI and breast MRI images. The outcome is calculated on the mean (M), standard deviation (STD), entropy (E) and accuracy (ACC). This algorithm is matched with adaptive K means approach. The proposed algorithm gives enhanced outputs as compared to existing approach.