Research on Image Segmentation Algorithm Based on Fuzzy Clustering and Spatial Pyramid

Abstract

An image segmentation algorithm based on fuzzy clustering and Spatial Pyramid is proposed in the paper. Firstly, the global color feature of the image is extracted. Secondly, the image under each scale is divided into several sub-blocks, and the feature vectors extracted from each sub-block are connected orderly. Thirdly, the saliency of each sub-block is expressed as the weight of the sub-block in the joint feature vector. Finally, the weighted eigenvectors are clustered. Experimental results show that accuracy and time complexity of the proposed algorithm are satisfactory.