Sub-pixel classification using FCM and FWCM algorithms

In this paper, the sub-pixel classification algorithms such as fuzzy c-means and fuzzy weighted c-means algorithms are implemented, which estimates the proportion of class components within a pixel. Experiments are carried out with the QuickBird multispectral image and the classification accuracy is estimated by the confusion matrices. The performance of these sub-pixel classification algorithms are evaluated and compared. The results compared indicate that the accuracy of fuzzy weighted c-means algorithm is more when compared to fuzzy c-means algorithm.