Machine Learning-Based Water Body Extraction from Satellite Images

Water body extraction from satellite images help us to study water resources. Due to constant changes in weather and increase of global warming it is necessary to track changes in water resources to study their impact on environment. From agriculture to industries, water bodies play an important role in production. Water bodies play an important role in keeping environmental balance. To study water bodies, acquiring high resolution satellite images is an important step. But even though after receiving high resolution images, there are some issues which are faced during extraction water bodies from satellite images. One of the most difficult tasks to overcome is to distinguish between water bodies and its surrounding shadows which are casted on the water. Detection of narrow water bodies is also challenging. Studying water bodies helps us in detection of floods, draught detection, and resource study and also how this water bodies can be used for developing water transport.