DA-Net: Dual Attention Network for Flood Forecasting

Abstract

Flood prediction is a challenging task due to the extreme runoff values, short duration, and complex generation mechanisms. This paper introduces DA-Net, a dual attention embedding network that incorporates convolution self-attention (CSA) and Temporal-related Feature Attention (TFA) to improve flood forecasting accuracy. CSA captures local context, while TFA enhances global feature modeling. The proposed method outperforms existing deep learning models on the Changhua and Tunxi watershed datasets.

Yirui Wu

Young Professor, CCF Senior Member

My research interests include Computer Vision, Artifical Intelligence, Multimedia Computing and Intelligent Water Conservancy.