An Improved Region Growing Algorithm for Image Segmentation

In this paper, we have made two improvements in region growing image segmentation. The First one is seeds select method, we use Harris corner detect theory to auto find growing seeds, through this method, we can improve the segmentation speed. The second one is growing rule. The homogeneity criterion usually depends on image formation properties that are not known to the user. We induced a new uncertainty theory-Cloud Model to realize automatic and adaptive segmentation threshold selecting, which considers the uncertainty of image and extracts concepts from characteristics of the region to be segmented like human being. Parameters of the homogeneity criterion are estimated from sample locations in the region. The method was tested for segmentation on general photo image and high-resolution remote sensing images. We found the method works reliable on homogeneity and region characteristics. Furthermore, the method is simple but robust; it can extract objects and boundary smoothly.