干旱区基于高光谱的棉花遥感估产研究

Study on Hyperspectral Remote Sensing Date of Cotton in Estimating Yield of Arid Region in China

【目的】揭示棉花产量与冠层光谱植被指数相关关系,建立棉花高光谱估算模型,促进高光谱技术在棉花长势监测和估产中应用。【方法】结合棉花生长发育规律,对棉花各时期冠层进行高光谱反射率测定,根据光谱曲线特征构建高光谱植被指数,基于棉花盛蕾期至吐絮后期7次地面光谱和产量测定,对光谱反射率与产量进行统计分析。【结果】各生育期可见光波段、近红外波段及短波红外波段光谱反射率与产量间分别达显著负相关、显著正相关与显著负相关水平。根据棉花冠层光谱波形特征,利用植被红边波段560nm反射峰、670nm吸收谷、近红外波段890nm反射峰、980nm和1210nm两个弱水汽吸收谷、短波红外1650nm和2200nm反射峰,设计归一化差值光谱指数,并与棉花产量进行相关分析,利用上述波段组合定义的归一化差值光谱指数与产量在各生育期均达显著或极显著相关,而VARI_700抗大气植被指数在各生育期均达极显著相关。【结论】以VARI_700抗大气植被指数建立各生育期的产量预报模型,为实现棉花营养生长期长势监测与产量预报提供依据。

[Objecive] The aims of this study were to analyze the correlation coefficients between yield and index of vegetation and red edge parameter of canopy, to build a high spectral model to estimate yield and to promote application of high spectrum technique in cotton. [ Method ] The correlation coefficients between yields and spectral reflectance at seven different growth stages were firstly calculated. According to spectrum curve characteristics index of high spectrum and vegetation was established. [ Result ] The statistical results show that yield is negatively correlated with spectral reflectance in visible light bands, positively correlated in NIR bands, negatively correlated in shortwave-infrared bands for all growth stages. Secondly, the normalized different spectral indices combined by [560 nm, 670 nm], [670 nm, 890 nm], [890 nm, 980 nm], [920 nm, 980 nm], [857 nm, 1 210 nm], [820 nm, 1 650 nm], and [820 nm, 2 200 nm] are significantly correlated with yield at all the seven growth stages, respectively, and VARI-700 is especial significantly correlated with yield at all growth stages. [ Conclusion ] Finally, the remote sensing models at different growth stages for cotton yield were built by VARI-700. It has provided a basis for monitoring cotton crop growing trend and forecasting production.