A NEW PICTURE FUZZY CLUSTERING METHOD TO SEGMENT THE SURFACE WATER FROM SATELLITE IMAGES

Abstract

In recent years, the research directions on advanced fuzzy sets have been received a lot of attention from scientists, typically studies on Picture Fuzzy Set (PFS). Picture fuzzy set was proposed to solve the problems of noisy data in order to improve the clustering performance.  With four membership degrees, including the positive, the neutral, the negative and the refusal, the optimal models using Picture fuzzy set have more choices and may yield more accurate results. In this paper, we propose a new Picture fuzzy clustering method named as PFSFCM. The proposed method is implemented and experimentally compared against the related methods, including the standard Picture fuzzy clustering (FCPFS), and the Confidence-weighted safe semi-supervised clustering (CS3FCM) in quality of clustering results with both of UCI dataset and Seattle Surface Water Dataset. The experimental results show that the proposed method has better performance comparing to selected methods on the same datasets.