Remotely Sensed Multispectral Satellite Image for Land Cover Classification

In an urban environment natural and human-induced environmental changes are of concern today
because of deterioration of environment and human health. The number of water bodies is declining
day by day whereas the concrete jungle is increasing. The study of land use and cover changes is
very important to have proper planning and utilization of natural resources and their management.
Remote sensing has become an important tool to develop and understand the global, physical
processes affecting the earth. In this chapter, we describe two machine learning algorithms which can
be used to analyse the change in the land cover for a period of over 10 years. We compare the
accuracies of the two algorithms applied to the same datasets and plot the same. We conclude by
stating that the performance of any algorithm depends on the dataset. It is very important to know the
characteristics of the input data before applying any classification process. The algorithms were
designed using Python programming language. The details about the dataset used, the relevant
equations, methodology and results have been provided in this chapter.