Classification and Evaluation of Land Cover Variations Using Landsat Data
The importance of the quantitative statistics related to earth’s natural resources and its land cover maps is undeniable. The data obtained from satellite platforms can offer a detailed overview of the natural processes and physical activities at incredible temporal and spatial resolutions. Mapping satellite images, classifying them and determining the vegetation area is a very important task for management and future planning of natural resources. The change detection practices are very important for discerning the spatial features and their conversion to urban lands. Over the years, the importance of timely and precise information presenting the extent and nature of natural land resources, especially in mountainous areas, has increased. In this paper, we have determined the vegetation cover and derived change maps using remotely sensed image data of the Abbottabad region of Pakistan. For this study, land cover mapping was achieved through interpreting satellite images of the region using ENVI 5.3 and ArcGIS Pro. Landsat images were utilized to estimate changes in land utilization patterns. To identify the changes between the years 1990 and 2016, the images were pre-processed and were categorized into five classes -i.e. forest, water bodies, settlements, agriculture land, and barren land using the maximum likelihood classifier. The post-classification change detection performed over classified images shows a positive change in forest class, water bodies, and settlements while a negative change in the barren land, with most of the barren land been converted into agricultural land. Some of the previously cultivated areas have been converted into the forest.
Copyright (c) 2020 Journal of Information Communication Technologies and Robotic Applications
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.