A Region-Based Image Segmentation Method with Mean-Shift Clustering Algorithm

Abstract

A method of region-based image segmentation with mean-shift clustering algorithm is introduced. This method first extracts color, texture, and location features from each pixel to form feature vector by selecting suitable color space. Then, these feature vectors are clustering with mean-shift clustering algorithm and the window parameter r is decided by the proposed method of selecting optimal clustering amount, so the numbers and the centers of clusters are also selected, and each pixel is grouped and labeled. Finally, the regions with the same label are segmented again according to the neighbor connection theory for pixels and a lot of the features which describe the regions are provided. Experiment results show this method can segment images quickly and has good segmentation results.