A new smoothness based strategy for semi-supervised atmospheric correction: application to the Léman-Baïkal campaign

The estimation of reflectance from hyperspectral data, a process called atmospheric correction, is a critical first step. Most of the methods need measurements of ground reflectances or atmosphere. When no information is available, unsupervised methods as Quick Atmospheric Correction (QUAC) exist assuming a large variability on the scene. But when studies on specific materials are conducted, no such large variability occurs. To solve the atmospheric correction in that case, we propose a new atmospheric correction method, called Smoothing Technique for Empirical Atmospheric Correction (STEAC) using the smooth property of the reflectance. This method is benchmarked against QUAC method on images acquired during the Leman-Baikal campaign. Results shows that this new method outperforms QUAC method, both in terms of accuracy and stability with respect to the scene heterogeneity.