Segmentation Google Earth Imagery Using K-Means Clustering and Normalized RGB Color Space

Image segmentation is defined as: "the search for homogenous regions in an image and later the classification of these regions". In this research, a remote sensing image, Pekanbaru city of Riau Province-Indonesia is provided for the green land segmentation. It is obtained by observing the surface of the earth using the Google Earth Imagery. To segment the green land of the given image, two different methods are used in this research, K-Means Clustering and Normalized RGB Color Space methods. This research is expected to have two clusters output: the spreading of green fields and not green fields. The result shows that the given Google Earth imagery can be segmented about 40.50 and 47.01 % pixels from all image pixels by K-Means Clustering and Normalized RGB Color Space respectively.