Abstract:?The extraction of sugarcane planting distribution is the basis of crop monitoring. It is of great significance for sugarcane monitoring and planting structure adjustment to extract sugarcane planting distribution timely and accurately. Based on Sentinel-1 and Sentinel-2 data, this paper adopts the active and passive remote sensing collaborative method and decision tree classification method to carry out research on the extraction method of sugarcane planting distribution in Guangxi, and combines with multi-year survey ground sample data for verification. On this basis, the data set of sugarcane planting distribution in Guangxi from 2018 to 2020 is extracted. The overall accuracy of sugarcane planting distribution in Guangxi in 2018 obtained by remote sensing extraction by this method is 92%, and the Kappa coefficient is 0.85; the overall accuracy of sugarcane planting distribution in Guangxi in 2019 obtained by this method is 94%, and the Kappa coefficient is 0.88; the overall accuracy of sugarcane planting distribution in Guangxi in 2020 obtained by this method is 94%, and the Kappa coefficient is 0.88. This data set can be used as the basic data for the analysis of temporal and spatial changes of sugarcane in Guangxi, and can also provide basic data support for the optimization and adjustment of sugarcane production management and planting structure in Guangxi.
Keywords:?Sugarcane;?Sentinel-2;?Active and passive remote sensing collaboration;?Decision tree classification;?Guangxi