基于PIE-Engine的甘蔗榨季砍收进度监测

Monitoring of Sugarcane Harvesting Progress Based on PIE-Engine During Milling Season

为了及时获得甘蔗砍收进度信息,提高甘蔗榨季砍收进程的信息化管理水平,基于国产遥感云计算平台PIEEngine,以广西壮族自治区来宾市兴宾区2020/2021榨季为例,使用甘蔗砍收前3个关键时点的Sentinel-2多光谱影像合成归一化植被指数(NDVI)时序数据集,根据当地主要作物NDVI走势差异,采用非监督分类提取甘蔗种植区域。在此基础上,利用Sentinel-1双极化合成孔径雷达(SAR)数据的重访周期短和全天候的特点,合成雷达植被指数(RVI)时序数据集,经变化检测获取不同甘蔗种植区域的近似砍收日期。在PIE-Engine平台上开发了县域甘蔗榨季砍收进度信息遥感在线监测程序。使用该方法提取的2020年兴宾区甘蔗种植面积接近多年稳定的种植面积,实现了兴宾区2020/2021榨季甘蔗砍收进度的全区与按乡镇统计分析和展示。研究表明,基于国产遥感云计算平台的甘蔗砍收监测方法切实可行。

In order to timely learn sugarcane harvesting progress information,improve information management level of sugarcane harvesting process in sugarcane milling season,based on domestic remote sensing cloud computing platform PIE-Engine,taking 2020/2021 milling season of sugarcane in Xingbin District,Laibin City,Guangxi as an example,NDVI time series dataset was generated by using Sentinel-2 multispectral images at three key time points before sugarcane harvest,according to the NDVI trend differences of main crops in the area,sugarcane planting areas were extracted by unsupervised classification.On this basis,the radar vegetation index(RVI)time series data set was synthesized by using characteristics of short revisit period and all-weather of Sentinel-1 dual-polarization SAR data,and approximate harvesting dates of different sugarcane planting areas were obtained through change detection.An on-line remote sensing monitoring program of sugarcane harvesting progress in county area was developed on pie-Engine platform.The sugarcane planting area in Xingbin district in 2020 extracted by this program was close to the stable planting area for many years.Statistical analysis and display of sugarcane harvesting progress in Xingbin district 2020/2021 milling season was realized as well.Study confirmed that sugarcane harvesting monitoring based on domestic remote sensing cloud computing platform was feasible.