无人机高光谱载荷性能交叉验证

Cross-validation of performance about UAV hyperspectral load

无人机定量遥感已成为当前遥感研究领域的热点之一,辐射定标是定量遥感的基础,在定量遥感研究中,首先需要确保无人机载荷性能的可靠性。基于辐射定标原理,以多光谱仪的固有波段为中心波长,设置高光谱仪对应波段范围,利用灰阶靶标,将高光谱影像中靶标的辐亮度值与多光谱仪的光谱响应函数进行卷积运算,得到多光谱各波段的等效入瞳辐亮度模拟值,从而拟合获得无人机多光谱载荷辐射定标系数;最后通过辐射定标系数验证分析及光谱曲线比对,最终实现无人机高光谱载荷性能交叉验证。结果表明:对于均匀地物如裸土,定标先后平均相对误差小于0.12;两次定标系数相对差异在±1%之内,且两次拟合相关系数r2分别在0.96和0.92以上,传感器线性响应度较高。实验结果可为相关多、高光谱遥感载荷性能分析提供参考。

The quantitative remote sensing of UAV has become one of the hotspots in the field of remote sensing research.Radiation calibration is the basis of quantitative remote sensing,in quantitative remote sensing research,it is the first necessary to ensure the reliability of UAV load performance.Based on the principle of radiometric calibration,set the wavelength range of the hyperspectral with the intrinsic band of the multispectral as the center wavelength.Based on the gray-scale targets,the radiance value of the target in the hyperspectral image is convoluted with the spectral response function of the multispectral image to obtain the equivalent enthalpy radiance simulation value of each band of the multispectral spectrum,and then obtain the multispectral load radiation calibration coefficient of the UAV.Finally,through the calibration of the radiation calibration coefficient and the comparison of the spectral curves,the cross-validation of the hyperspectral load performance of the UAV is realized.The results show that the average relative error of calibration is less than 0.12 for the bare soil.The relative difference between the two calibration coefficients is between±1%,and the two fitting determination coefficients are above 0.96 and 0.92 respectively.The experimental results can provide reference for the correlation analysis of multispectral and hyperspectral remote sensing loads.