ABSTRACT
The process of supervised classification requires image analyst input based on direct knowledge of the areas on the ground or other much higher resolution airborne or satellite imagery that are to be classified within the imagery. These are represented by training sites and may be collected in the form of delineated polygons or representative pixels that will use to develop a multiband classification. Statistical algorithms are then used to analyze the spectral bands of the imagery and determine how closely each relates to the identified training samples representing the categories of interest throughout the entire image data. This exercise will explore the use of user-collected training sites and a maximum likelihood statistical algorithm for creating a supervised image classification to classify land use and land cover. Maximum likelihood classifier is the most widely used supervised classifier by users worldwide.
