Dynamic Texture Segmentation Using Robust Dynamic Fuzzy Clustering with Student's t‐Hidden Markov Model

The hidden Markov model (HMM), as a common method for sequential data modeling, has been applied to describe the dynamic texture. In this paper, we propose a robust dynamic fuzzy clustering method based on the student's t‐hidden Markov model for dynamic texture segmentation. First, we present a novel dissimilarity function between the observation and the cluster prototype, which is measured by the student's t‐hidden Markov model. Second, the model parameters of robust dynamic fuzzy clustering are derived by minimizing the objective function. Finally, robust dynamic fuzzy clustering is applied to dynamic texture segmentation. The experimental results on synthetic dynamic texture segmentation dataset and nature color image sequences demonstrate the efficiency of the proposed approach, which outperforms state‐of‐the‐art dynamic texture segmentation algorithms. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.