Fig. 11From: Atmospheric correction of vegetation reflectance with simulation-trained deep learning for ground-based hyperspectral remote sensingPerformance metrics of the testing set for models trained and tested with spectral radiance and reflectance at systematically reduced spectral resolution and increased FWHM. Noting the range of the spectral channels on the x-axis are presented in logarithmic scale, the model’s accuracy remains consistently \(>90\%\) down to a resolution of 29 channels (width of 15.5 nm). Accuracy lower than 80% occurs only at the lowest possible resolution of 2 channelsBack to article page