基于高光谱遥感的作物光合参量反演方法研究进展

Research Progress of Vegetation Photosynthetic Parameter Inversion Method Based on Hyperspectral Remote Sensing

光合作用对作物的生长发育、产量的形成及干物质的积累起着至关重要的作用。高光谱遥感反演技术凭借其准确快速和不破坏植被的优势,现已成为作物植被光合数据参量定量分析的重要方法。作物光合参数的定量遥感反演不仅能够了解作物的生长发育状况,还可为基于遥感的生态系统反演模型提供参考数据。研究综述了近年来科学文献中高光谱光合反演的研究成果,主要介绍了作物光合参数高光谱反演技术及研究现状,包括高光谱数据预处理、光谱变换、特征波段筛选、光谱植被指数、线性回归模型、偏最小二乘回归(PLSR)、机器学习模型、BP神经网络建模和支持向量机(SVM)回归模型等,并在此基础上,对作物光合高光谱反演中存在的问题进行了探讨。

Photosynthesis plays a crucial role in crop growth and development,yield formation and dry matter accumulation.Hyperspectral remote sensing inversion technology has become an important method for quantitative analysis of photosynthetic data parameters of crop vegetation due to its advantages of accuracy,speed and non-destruction of vegetation.Quantitative remote sensing inversion of crop photosynthetic parameters can not only understand the growth and development of crops,but also provide reference data for ecosystem inversion models based on remote sensing.This study reviewed the research results of hyperspectral photosynthesis inversion in recent years in scientific literature,mainly introduced the hyperspectral inversion technology of crop photosynthetic parameters and the research status,including hyperspectral data preprocessing,spectral transformation,feature band screening,spectral vegetation index,linear regression model,partial least squares regression(PLSR),machine learning model,BP neural network model and support vector machine(SVM)regression model,etc.,and on this basis,the problems existing in crop photosynthetic hyperspectral inversion are discussed.