Supervised Maximum Likelihood Approach for Land Use Land Cover Dynamics of Semi-Urban Region

Remote sensing offers a unique environmental monitoring capability that covers extensive geographical areas while capturing irreplaceable information on the Earth’s land, atmosphere and oceans. RS classifications of an image are the procedure of converting remotely sensed imageries to practical products. LU/LC classification using remotely sensing satellite data is significant research and generally used in RS applications. Nowadays, Remote Sensing technology has become one of the promising tools to measure the LU/LC changes. Data collection of two different time periods 2005 and 2017 are taken from Google Earth Image and are downloaded using Smart GIS 2019. In the pre-processing stage Geo-Referencing, subsetting is implemented using an ERDAS version 9.2. For the data 2005, about 76.19 percent of accuracy is obtained, for the data 2017, about 77.24 percent of accuracy is obtained. From 2005- 2017 there is a significant decrease in Grassland, Open land, Water body and Trees, there is a significant increase in the built-up, Highway and Tiled roof. About 33.85 percent of changes in land use are identified from 2005-2017 and about 25.06 percent of changes in the land cover is identified from 2005-2017