A Region-Based Image Segmentation Approach with KMC Algorithm of Automatic Determination K

In this paper, a K-means clustering (KMC) algorithm of automation determination the clustering number K is proposed, and a approach of region-based image segmentation is introduced based on our proposed algorithm. For this approach, firstly, a suitable color space is selected, the features of color, texture, and location are extracted, and the feature space is generated. Then, in this feature space, an image is clustered and separated into some regions by proposed method. Finally, the features of regions are extracted. The proposed approaches and algorithm are described in detail. Experiment results show the proposed approach has the quick segmentation speed and good segmentation results, and it is fit for region-based image retrieval system and has the better application value.