欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2005, Vol. 21 ›› Issue (1): 318-318.

所属专题： 油料作物； 玉米； 烟草种植与生产

• 目次 • 上一篇    下一篇

玉米和大豆LAI高光谱遥感估算模型研究

宋开山,张 柏,王宗明,段洪涛   

  • 出版日期:2005-01-05 发布日期:2005-01-05

Study on Hyperstectral Remote Sensing Estimation Model for the Corn LAI with Canopy Reflectance

Song Kaishan, Zhang Bai, Wang Zongming, Dunan Hongtao   

  • Online:2005-01-05 Published:2005-01-05

PDF (PC)

920

摘要： 以ASD FieldSpec光谱仪实测了不同生长季的大田玉米、大豆的冠层高光谱与作物的叶面积指数LAI。采用单变量线性与非线性拟合和逐步回归分析的方式，建立了玉米、大豆LAI高光谱遥感估算模型，并对模型的估算结果进行了初步分析。分析结果表明，绿光波段反射峰区、红光波段以及近红外区的单波段反射率与作物的LAI有较强的相关性，而其他波段的反射率与作物的LAI的相关性相对较弱；以高光谱的窄波段构造的NDVI和RVI与作物的LAI的相关程度高，回归模型的预测水平高；而以多波段逐步回归方式构造的统计模型的预测效果最好。

Abstract: Canopy reflectances of corns soybean are measured with ASD FieldSpec during different corn growing stages; and simultaneously, LAIs are acquired. The relationship between canopy reflectance and LAI are analyzed with single variant and multivariant regression. It found that reflectance in green waveband red waveband and near infrared waveband have an direct relation to corn LAI while reflectance in other wavelength have an loose relation to corn LAI growing in different stages. By our study, It found that NDVI and RVI constructed by hyperspectral remote sensing reflectance can greatly improve remote sensing ability for estimation corn LAI. By comparison, it can be drawn that multivariant regression is the best method for estimating corn LAI with remote sensing model.