Accurate information on urban surface water is important for assessing the role it plays in urban ecosystem services in the context of human survival and climate change. The precise extraction of urban water bodies from images is of great significance for urban planning and socioeconomic development. Deep learning is that learning method that recreates the human mind. It will precisely separate elevated level choices from low-level choices of the info picture . We have proposed a completely unique deep-learning architecture to extract various earth objects from medium resolution satellite imagery using deep learning approaches. We have used the Convolutional Encoder – Decoder model on the Sentinel -2 Satellite images and Experimental results show that the proposed method achieved higher accuracy for water feature extraction with average overall accuracy is 99.9%.