Abstract

Abstract
Recent advances in the field of aerospace technology have included the collections of satellite images data in a huge number. With this data will be easy to study about the environmental change in the specific area by monitoring the land use and land cover. The most of land use changes are caused by human activities, such as cutting the forest for agriculture land or for urban area.

Clustering as an unsupervised classification method can be used to obtain some initial information from satellite images data. This method is based on partitioning a collection of data points into a number of subgroups, where the objects inside a cluster (a subgroup) show a certain degree of closeness or similarity. However, classifying urban areas using satellite data or multispectral data is difficult due to several different types of land cover.

One solution is to apply fuzzy clustering method. By conventional clustering methods, class is either assigned to or not assigned to a defined group. Fuzzy clustering which apply the concept of fuzzy sets to cluster analysis may pertaining to group at each pixel of the images data by a membership function which associated to each cluster ranging between 0 and 1.

In this study, a fuzzy clustering method that we implemented in statistical software XploRe is used to classify multi-spectral Landsat TM images. The result of this investigation shows that the underlying structures and patterns from satellite images data can be classify more precisely compare to conventional ones.