Novel initialization scheme for Fuzzy C-Means algorithm on color image segmentation
Applied Soft Computing
| Hi-index | 0.00 |
Fuzzy c-means algorithm with spatial constraints (FCM_S) is more effective for image segmentation. However, it still lacks enough robustness to noise and outliers, and costs much time in computation. To overcome the above problem, a new algorithm for image segmentation based on fast fuzzy c-means clustering is proposed in this paper. In order to reduce the number of iteration, the algorithm selects the peak value of gray histogram as the initial centroid. To enhance the noise immunity, the clustering of centre pixel is influenced by the neighbor mean value and median value. The algorithm reduces the time of each iteration step by the gray histogram of image. The experimental results on two types of images indicate that the proposed algorithm is effective and efficient.