We propose a method for eliminating the atmospheric effect in polarimetric imaging remote sensing by using polarimetric imagers to simultaneously detect ground targets and skylight, which does not need calibrated targets. In addition, calculation efficiencies are improved by the skylight division method without losing estimation accuracy. Outdoor experiments are performed to obtain the polarimetric bidirectional reflectance distribution functions of painted surfaces and skylight under different weather conditions. Finally, the roughness of the painted surfaces is estimated. We find that the estimation accuracy with the proposed method is 6% on cloudy weather, while it is 30.72% without atmospheric effect elimination.

中文翻译：

---

改进的大气效应消除方法，用于评估涂漆表面的粗糙度

我们提出了一种通过使用偏振成像仪同时检测地面目标和不需要校准目标的天窗来消除偏振成像遥感中的大气影响的方法。另外，通过天窗划分方法提高了计算效率，而不会损失估计精度。进行户外实验以获得在不同天气条件下涂漆表面和天窗的偏振双向反射率分布函数。最后，估计涂漆表面的粗糙度。我们发现，在多云天气下，该方法的估计精度为6％，而没有消除大气影响的估计精度为30.72％。