With the continuous development of science and technology level, remote sensing technology has a wide range of application prospects in water pollution monitoring. The study combines the Grab-Cut image segmentation algorithm, atmospheric radiation correction, and water pollution identification to construct a remote sensing technology-based method for monitoring water pollution. The research is focused on analyzing the water quality distribution pattern of the upper and lower lakes in the sample lake area using the constructed water pollution monitoring method. On this basis, from the four dimensions of water pollution prevention and control, recycled water recycling, ecological restoration and protection, and environmental monitoring, the comprehensive management measures for water pollution are proposed. The pollutant contents of the upper and lower lakes before and after the management are compared to explore the effect of the comprehensive management of water pollution. The results show that COD contributes 86.77% to the water pollution in the sample lake area. The water quality of the upper and lower lakes is more distributed by class V and class III, which account for 36.67% and 43.33%, respectively. The proposed water pollution monitoring method is able to accurately identify and classify the pollution. After the comprehensive treatment, the COD and ammonia nitrogen content of the upper lake decreased by 30.51% and 37.43%, and that of the lower lake decreased by 35.90% and 39.06%. The effect of water pollution treatment was remarkable.