ABSTRACT
The operation of grouping or sorting pixels within an image based on the p spectral values, into various categories is known as image classification. The unsupervised method of classification typically requires little or no input from the image analyst in developing the output land-use/land cover image classification. This classification system uses statistical means and covariance matrices to iteratively assign each pixel to a designated output class based on how spectrally separate each group of clustered pixels are. This exercise explores the general procedures for image classification using the unsupervised image analysis process to classify land use and land cover.
