摘要
针对目前高原地区青稞种植产业的实时监测困难、估产成本过高的问题,提出了一种依托遥感手段的新方法。基于S-G滤波去噪后的时间序列MODIS-NDVI数据与气象数据,实现了青稞种植面积精准提取和长势监测,并利用模型分解法构建了日喀则地区的青稞产量预测模型。结果表明,青稞单产预测模型的平均估产误差为7.53%,平均相关系数为0.9426。该方法可为高原地区青稞种面积的快速准确提取以及青稞产量预测提供参考,且具有较好的操作性和适用性。
A new method based on remote sensing is presented for the problems that difficulty in real-time monitoring and high cost of estimating the production of highland barley planting industry in the plateau area.In the study,the planting area of highland barley is accurately extracted and its growth is monitored by using the time series MODIS-NDVI data and meteorological data after S-G filtering and denoising based on remote sensing technique,and a highland barley production forecast model in the Rikaze is built by using decomposition method.The result shows that average yield estimation of highland barley yield estimation error is 7.53%,and the average correlation coefficient is 0.9426.The proposed method has good operability and applicability for the rapid and accurate extraction of barley species area and the forecast of barley yield in plateau areas.
作者
潘莹
仙巍
陈春容
陈军
徐薇
PAN Ying;XIAN Wei;CHEN Chunrong;CHEN Jun;XU Wei(Institute of Resources and Environment,Chengdu University of Information and Technology,Chengdu 610225,China;Rikaze Meteorological Bureau,Rikaze,Tibet 857000,China)
出处
《遥感信息》
CSCD
北大核心
2022年第2期77-83,共7页
Remote Sensing Information
基金
四川省科技计划项目(2020YFG0146)
西藏自治区自然科学基金项目(XZ2019ZRG-104(Z))。
关键词
青稞
日喀则地区
时间序列
模型分解
遥感气象估产
highland barley
Rikaze
time series
model decomposition
meteorological production estimation by remote sensing
