Land Use/Land Cover Classification over Dachigam National Park, Jammu & Kashmir Using Supervised Artificial Neural Networks and Support Vector Machines
Abstract
Land use/land cover (LULC) information has a wide range of applications. Satellite images are an important tool for LULC analysis providing global data used to detect land cover changes and plays a vital role in assessing the changes in ecological sensitive landforms such as glaciers, forests, geological formations, and many more. The literature domain cites various algorithms for this purpose using remotely sensed images. The main objective of this study is focused on monitoring LULC change detection over Dachigam National Park, (Jammu & Kashmir) for the period of 30 years, (1989-2019). In this paper, we used multispectral and multi-temporal Landsat data and Supervised Image Classification techniques such as; Artificial Neural Network (ANN) Support Vector Machine (SVM). The results showed better performance of SVM in land cover classification than ANN. Our findings conclude significant changes in all the selected land cover classes used for the classification also impact the climatic change and its variability. Therefore, the final analysis of the study is used for better monitoring and development through strategic planning for restoration and design an effective sustainable management plan for the National Park.
