A fuzzy clustering algorithm with spatial robust estimation constraint for noisy color image segmentation

In this paper, we introduce two enhanced Fuzzy C-Means (FCM) clustering algorithms with spatial constraints for noisy color image segmentation. The Rank M-type L (RM-L) and L-estimators are used to obtain the sufficiently spatial information of the pixels. These estimators are involved into the FCM algorithm to provide robustness for the proposed segmentation schemes. The performance of the proposed algorithms is tested in real images under different noise conditions by simulating salt and pepper, Gaussian, and speckle noises, as well as with two mixtures of them. Simulation results indicate that the proposed methods consistently outperform other color image segmentation algorithms used as comparative. Additionally, the proposed algorithms are tested for segmenting a remote sensing image, where the noise is not known beforehand implied. Finally, the proposed algorithms have the robustness and effectiveness needed for image segmentation in the presence and absence of noise.

► We present a novel FCM with spatial constraints for noisy color image segmentation.
► The Rank M-type L-estimator is used to obtain the sufficiently spatial information.
► The performance of algorithm is tested under different conditions and types of noise.
► The proposed method outperforms other color image segmentation algorithms.