Main Article Content

M.R. Thiyagupriyadharsan
S. Suja

Abstract

This paper presents the clustering techniques fuzzy C- means clustering(FCM) and subtractive clustering for Hyperspectral image dataset. The FCM should be done only on known number of clusters and subtractive clustering can be done on unknown data also. First FCM is done on known hyperspectral image dataset and for the dataset the subtractive clustering is done. Both the algorithms yielded good results. Then both the algorithms performance are evaluated by using Fuzzy Inference system (FIS) model. The model shows improvement in the test data over the training data.

Article Details

References

[1] Bezdek, FCM The fuzzy cmeans cluster ing algorithm, IEEE