Color Image Segmentation Algorithm Based on Affinity Propagation Clustering

摘要: This paper proposes a color image segmentation algorithm based on affinity propagation clustering. Firstly, the is converted to other space; and then sampled, sampling data clustered by with given number, we can get their class label of rest according maximum similarity rules; finally, using regional combined methods ,we remove independent small regions, revised result. The simulation experiment shows that this method achieve application in large-scale segmentation. And it has faster processing speed satisfactory results, more line human whole visual perception.