 |
Fast Robust Fuzzy Clustering Algorithm for Grayscale Image Segmentation
1 : Laboratoire des technologies innovantes
(LTI)
* : Auteur correspondant
Image segmentation is a primordial step in the chain of image processing. A wide range of image segmentation methods are based on the FCM algorithm. However, the lack of any spatial information in this latter algorithm makes it very sensitive in the presence of noise. To overcome this problem and faster the segmentation process, we propose two versions of a fast and robust extension of the FCM algorithm.
To prove the strength of the proposed algorithms we compare them, in a qualitative and quantitative standpoints, and four other fuzzy clustering algorithms. To this end, we use synthetic and real images as testing data.
