The water body on the earth’s surface is an important element of the environmental ecosystem. Floating matter, suspended matter, and dissolved matter negatively affect most traditional methods used to extract water body information from remotely sensed images. As a result, extracting water body information with high precision from a wide range from remote sensing images that contain complex ground-based objects has proved difficult. In this study, we proposed a method of extracting water body information from a remote sensing image that considers the wetness of ground-based objects, and we carried out threshold value selection using the Otsu method and post-processing supported by mathematical morphology. Hangzhou Bay, which is rich in inland water resources and marine resources, was selected as the study area. The results obtained from a Landsat 8 Operational Land Imager (OLI) image show that the contours and spatial extent of the water extracted by the method are highly consistent, with reduced effects of floating matter, suspended matter, and dissolved matter; the producer’s accuracy and user’s accuracy are 0.9706 and 0.9572, respectively. Overall, the proposed method can provide technical support for the accurate extraction of water body information, which is of great significance for the scientific management of water resources.