Super-resolution reconstruction of sea surface pollutant diffusion images based on deep learning models: a case study of thermal discharge from a coastal power plant

While remote sensing images could convey essential information of surface water environment, the low spatial resolution limits their application. This study carried out a series experiment tests thermal discharge from coastal power plant and constructed an image dataset HY_IRS, representing transport diffusion discharged heated in tidal waters. Two super-resolution (SR) reconstruction models based on deep learning (DL), ESPCN ESRGAN, were trained this then used to reconstruct high-resolution images. It shows that two DL can markedly improve discharging, with PSNR improved by 8.3% average. The successfully infrared SST Landsat8 TIRS, indicating SR model has good effect crucial application prospect field improving pollutant