ABSTRACT
Land use/land cover (LULC) classification from remote sensing data is a complex procedure. Many factors such as the spatial resolution of the remotely sensed data, availability of different data sources, a suitable land cover classification system, availability of image-processing software, use of a suitable classification algorithm, and the analyst’s experience may all affect the classification results (Lu and Weng 2007). The major steps for LULC classification from remotely sensed data have been discussed by Lu and Weng (2007). For a specific LULC classification, two critical questions need to be answered: What kinds of remote sensing data and variables are used? What algorithm is suitable for this kind of dataset for LULC classification?
