Envisioning Earth: Enabling Environmental Monitoring with U-Net Segmentation of Forests and Waterbodies in Satellite Imagery

Inspired by the pressing need for addressing environmental issues like water scarcity and deforestation, this study uses machine learning and image processing on satellite data to detect changes comprehensively. Our method focuses on precisely separating forests and water bodies using Convolutional Neural Networks (CNNs), more especially the U-Net model. This allows for accurate tracking of environmental changes. The use of cutting-edge segmentation algorithms, dataset analysis, and literature evaluation are all components of the methodological rigor. The approved result shows how effective deep learning is at monitoring the environment and offers insightful information for sustainable management and to make informed decisions. The aim of this paper is to add a methodical methodology while highlighting deep learning's transformative potential to improve environmental conservation efforts and obtained accuracy of 0.840 and 0.880 for Forest Segmentation and Water Body Segmentation respectively