Black and odorous water detection of gaofen-2 remote sensing images based on deep learning

Black and odorous water seriously affects the ecological balance of rivers and the health of people. Satellite remote sensing technology with its advantages of large range, long time series, low cost, and high efficiency, has provided a new area for water quality detection. In this paper, Gaofen-2 remote sensing data with a spatial resolution of 1 m is leveraged as the data source. We build a high-quality remote sensing image dataset to enrich the data source in the northern coastal zone of China. In addition, we propose a network with an encoder-decoder discriminant structure for black and odorous water detection. In the network, an augmented attention module is designed to capture more comprehensive black and odorous water semantic information. Further, the new loss function is adopted to solve the class imbalance. Experimental results demonstrate that the network is superior to other state-of-theart semantic segmentation methods on our dataset.