Novel coupled DP system for fuzzy C-means clustering and image segmentation

This study proposed a novel fuzzy c-means clustering method which calculates the density of the data points based on the weighted mean distance (WMFCM). A novel coupled DNA-GA and P system (DP system) is introduced to realize the clustering process. The evolutional rules in the coupled DP system can help the WMFCM algorithm jump out of the local optimum and get the initial clustering centers. The performance of the coupled DP system in dealing with fuzzy clustering problems is measured by conducting experimental analysis on UCI datasets and BSDS300 image datasets. And the experimental results are compared with several popular algorithms. Experimental results show that our algorithm can perform better than other algorithms.