Flood hazard assessment model based on ST-UNet

Abstract

Frequent flooding disasters occur in China, including a number of very heavy rainfall events have brought great losses to the country and the people, and these sudden extreme weather events seriously threaten the lives and property safety of the people. The main research of this paper includes the stitching and processing technology of UAV images, the research and use of deep learning model, and the establishment of flood disaster assessment model. In this paper, a flood disaster assessment model based on ST-UNet and fuzzy integrated evaluation is proposed and experimentally validated. The results show that the model has high accuracy and practicality, and can provide scientific basis for flood warning and defense decision. The research results of this paper are of great practical significance and application value for strengthening flood disaster warning and emergency response capability. CCS CONCEPTS•Computing methodologies∼Modeling and simulation∼Simulation evaluation