The paper "Extraction of Surface Water Bodies using Optical Remote Sensing Images: A Review" by R Nagaraj and Lakshmi Sutha Kumar critically examines the methods used for extracting surface water bodies (SWB) from optical remote sensing images. The authors highlight the importance of SWM in water resource management, environmental conservation, and disaster preparedness, especially in the context of rapid urbanization, overutilization, and environmental degradation. The review covers various techniques, including Spectral Indices (SI), Machine Learning (ML), Deep Learning (DL), and Spectral unmixing, with a focus on satellite data, study areas, methodologies, results, advantages, and disadvantages. The study reveals that DL outperforms ML and SI due to its advanced network design elements like skip connections, dilation convolution, attention mechanisms, and residual blocks. Spectral unmixing addresses mixed pixel misclassification issues. The paper discusses the implementation and results of these methods and suggests future directions, such as integrating DL with spectral unmixing, fusing multisource data (SAR and optical), and incorporating ancillary data (DEM). The introduction emphasizes the significance of accurate water body extraction for effective water management, highlighting the limitations of traditional methods and the transformative role of remote sensing technologies. The paper also provides a brief overview of the sensors used in SWB extraction and the critical aspects of optical sensors, such as spatial, temporal, spectral, and radiometric resolution.The paper "Extraction of Surface Water Bodies using Optical Remote Sensing Images: A Review" by R Nagaraj and Lakshmi Sutha Kumar critically examines the methods used for extracting surface water bodies (SWB) from optical remote sensing images. The authors highlight the importance of SWM in water resource management, environmental conservation, and disaster preparedness, especially in the context of rapid urbanization, overutilization, and environmental degradation. The review covers various techniques, including Spectral Indices (SI), Machine Learning (ML), Deep Learning (DL), and Spectral unmixing, with a focus on satellite data, study areas, methodologies, results, advantages, and disadvantages. The study reveals that DL outperforms ML and SI due to its advanced network design elements like skip connections, dilation convolution, attention mechanisms, and residual blocks. Spectral unmixing addresses mixed pixel misclassification issues. The paper discusses the implementation and results of these methods and suggests future directions, such as integrating DL with spectral unmixing, fusing multisource data (SAR and optical), and incorporating ancillary data (DEM). The introduction emphasizes the significance of accurate water body extraction for effective water management, highlighting the limitations of traditional methods and the transformative role of remote sensing technologies. The paper also provides a brief overview of the sensors used in SWB extraction and the critical aspects of optical sensors, such as spatial, temporal, spectral, and radiometric resolution.