 Our research has shown that using BRDF signatures can improve the accuracy of crop mapping when compared to using reflectance data from a single nadir observation direction. This is because BRDF signatures provide more detailed information about the surface properties of crops, which can lead to better classification results. Additionally, our study showed that the random forest classifier was the most accurate, followed by classification and regression trees, support vector machines, and then naive base. This article was authored by Jijunjin, Shinguachen, Tian Gangin, and others.