DEEP LEARNING APPROACH FOR FLOOD DETECTION USING SAR IMAGE: A CASE STUDY IN XINXIANG

Abstract. With the gradual warming of global climate, frequent floods have caused huge losses to human life and property. Flood mapping by SAR image has been an important topic, it is increasingly use deep learning method extract flood information. In order achieve automatic extent extraction, this paper proposes attention mechanism-based water body extraction network with GF3 images, successfully for detection in Xinxiang, Henan, China. paper, proposed incorporates channel mechanism position based on U-Net, improve efficiency accuracy ignore unimportant information weight make model focus toward related body. The OA our can reach 0.959 Recall 0.942 verifying four sets test data. Experiments show fast Xinxiang be achieved method.