Exploration of River systems using Artificial Intelligence techniques

River systems are extensively watched and scrutinized for a variety of objectives, including estimating the geographical placements of water bodies, monitoring river gauge levels, project flows of river, and saving sustainable energy supplies as a result of global warming. Various segmentation and machine learning integration model are used to explore these river systems on digital imagery systems. Different datasets from Google Earth Engine and Kaggle are used in this paper. In order to extract river systems from satellite pictures, segmentation techniques such image segmentation, gray scaling, enhancement, global thresholding, and Deep Learning UNet Architecture are merged. This leads to an 81.02% dice score for the constructed UNet Model. Thus, rivers can be extracted from satellite photos using these newly established algorithms. This is applicable to different datasets for semantic segmentation identification.