基于热红外遥感影像的玉米田间土壤水分反演研究  

Inversion of soil moisture in corn field based on thermal infrared remote sensing image

作　　者：郭辉[1] 卜小东[1] 黄可京[1] 刘英超[1] Guo Hui;Bu Xiaodong;Huang Kejing;Liu Yingchao(Beijing Vocational College of Agriculture,Beijing 102208,China)

机构地区：[1]北京农业职业学院,北京市102208

出　　处：《中国农机化学报》2020年第10期203-210,共8页Journal of Chinese Agricultural Mechanization

基　　金：北京市教育委员会科技资助项目(KM201812448006)。

摘　　要：土壤水分监测对掌握农作物的生长状态至关重要。本研究为了在玉米作物的主要生育期有效地反演田间土壤含水量。本文以无人机平台获取的热红外遥感影像作为数据源,基于热惯量法反演田块尺度的土壤含水量。通过建立土壤热惯量与土壤含水量之间的线性回归模型,在试验田进行模型精度验证。结果表明,在实际农田环境中基于热惯量方法反演土壤含水量时,随着灌溉水平的提高其反演精度先升高后下降。模型在不同灌溉水平下反演土壤含水量的精度验证结果为:R^2=0.71,RMSE=3.09%。热惯量法具有较高的土壤含水量反演精度,为基于无人机热红外遥感田间土壤含水量监测提供了参考。Soil moisture monitoring is essential for tracking crop growth.The purpose of this study is to effectively retrieve the field soil water content during the main growth period of the corn.It was used thermal infrared remote sensing images acquired by the drone platform as a data source,and was based on the thermal inertia method to retrieve soil moisture at the field scale.By establishing a linear regression model between soil thermal inertia and soil water content,it performed model accuracy verification in test fields.The results show that when the soil moisture content is retrieved based on the thermal inertia method in an actual farmland environment,as the water treatment level increases,its inversion accuracy increases first and then decreases.The accuracy verification results of the model inversion of soil water content under different water treatments are:R^2=0.71,RMSE=3.09%.The research results show that the thermal inertia method has higher accuracy of soil moisture retrieval,which provides a reference for monitoring soil moisture in the field based on UAV thermal infrared remote sensing.

关 键 词：土壤水分 玉米 热红外遥感 反演 

分 类 号：S275[农业科学—农业水土工程] TP79[农业科学—农业工程]