基于高光谱红边参数的棉花冠层覆盖度提取研究

Study on Extracting Cotton Canopy cover with Hyperspectral Red-edge Parameters

利用ASD FieldSpec FR野外光谱仪,获取了棉花(2个品种4种种植方式)各生育时期的反射光谱数据,同时,用数码相机测试了棉花冠层覆盖度;研究了棉花冠层光谱红边参数随作物生育期的变化规律,并进行了棉花光谱数据和冠层覆盖度的相关回归分析。结果表明:归一化植被指数(NDVI)与冠层覆盖度的线性相关达到1%极显著水平(r=0.6303**,n=62),可以利用NDVI反演棉花冠层覆盖度;用红边斜率建立的二次函数模型的相关系数最高(r=0.6983**,n=62),总均方根差RMSE为0.0476g/m2能较好地提取棉花冠层覆盖度,说明可为生产上利用遥感手段大面积、无破坏、及时评价棉花生长状况提供重要依据。

Using the ASD Fieldspec FR spectrometer, get hyperzpectral date of cotton （two varieties, four planting patterns） at key growth stages,at the same time, obtain the cotton canopy cover by using digital camera; We analyse the regular pattern of red - edge along with the different growth stages. The results showed that the significantly correlation between NDVI and cotton canopy cover（ r = 0. 6303 ＊ ＊ , n = 62）. So we can using NDVI to estimate cotton canopy cover ; The slope of red - edge quadratic model has the highest correlation coefficient （ r = 0. 6983 ＊ ＊ , n = 62） . It is feasible for using RMSE of extracting cotton canopy cover（ RMSE = 0. 0476 g/m^2 ）. The study indicates that it can play a vital role in providing time - specific and time - critical information by using hyperspectral remote sensing for precision farming.