Monitoring the Land Use, Land Cover Changes of Roorkee Region (Uttarakhand, India) Using Machine Learning Techniques

Abstract

Satellite images play an important role for capturing Earth's surface. Using satellite images land cover monitoring could be done through which the modification or changes on land surface could be identified. Comparison can be made on the basis of past satellite image analysis, which helps to identify the changes that are occurring or have already occurred. Although there exist many techniques for land cover monitoring, proper land cover identification and detection of changes on the land cover is still a challenge. In the recent years, machine learning techniques have been utilized in distinct areas of image analysis and resulted in positive outcomes. Hence, in this paper, four supervised machine learning algorithms (i.e., support vector machine [SVM]), neural network [NN], maximum likelihood [MLC], and parallelepiped [PP] algorithms) have been utilized for land cover identification and detecting the amount of changes that have occurred in the individual land cover classes.