An interval type-2 fuzzy c-means algorithm based on spatial information for image segmentation

Fuzzy c-means algorithm (FCM) is a classic algorithm used in image segmentation. However, FCM is founded with type-1 fuzzy sets, which cannot handle the uncertainties existing in images and algorithm itself. The interval type-2 fuzzy c-means algorithm (IT2FCM) has better performance on handling uncertainties. But for image segmentation, IT2FCM hasn´t taken the spatial information of images into account, which makes the segmentation result not ideal enough. In order to incorporate spatial information, an extension of IT2FCM is proposed here. And the result of image segmentation using the proposed algorithm shows that the algorithm has better performance on suppressing noise and better effects on segmenting images compared with FCM-based algorithms and IT2FCM.