基于场景的热红外高光谱数据光谱定标

发布时间：2018-04-04 03:51

本文选题：热红外高光谱数据　切入点：光谱定标　出处：《红外与激光工程》2017年01期

【摘要】：传感器每个波段的中心波长和半高全宽(Full Width at Half Maximum,FWHM)随成像环境变化会发生较大的系统性漂移。这种漂移最终会影响发射率和温度的反演精度,尤其是在大气吸收波段附近的发射率反演精度。选择水汽在11.73μm处的吸收通道作为参考波段,提出了适用于热红外高光谱数据的光谱定标技术流程。模拟实验表明:光谱分辨率为50 nm,中心波长偏移在-50~50 nm、FWHM变化在-25~25 nm时,大气水汽含量对光谱定标误差的影响最大。同时,对误差分布曲面进行拟合得到描述误差分布模型,用于误差的估计。当大气水汽含量足够大时,光谱中心波长偏移估算误差可达到1 nm以内。最后,将所提方法应用于机载热红外高光谱数据光谱定标。结果显示,热红外高光谱成像仪中心波长偏移为28.4 nm,FWHM变化为-18.5 nm。
[Abstract]:The full Width at Half maximum FWHM of each band of the sensor has a large systematic drift with the change of the imaging environment.This drift will ultimately affect the accuracy of emissivity and temperature inversion, especially near the atmospheric absorption band.The absorption channel of water vapor at 11.73 渭 m is selected as the reference band, and the spectral calibration process suitable for thermal infrared hyperspectral data is proposed.The simulation results show that when the spectral resolution is 50 nm and the center wavelength shift is -50 nm FWHM varies from -25 nm to 25 nm, the atmospheric water vapor content has the greatest effect on the spectral calibration error.At the same time, the error distribution model is obtained by fitting the error distribution surface, which is used to estimate the error.When the atmospheric water vapor content is large enough, the estimation error of spectral center wavelength shift can reach less than 1 nm.Finally, the proposed method is applied to the spectral calibration of airborne thermal infrared hyperspectral data.The results show that the central wavelength shift of the thermal infrared hyperspectral imager is 28.4 nm and FWHM changes to -18.5 nm.
【作者单位】：中国科学院上海技术物理研究所中国科学院空间主动光电技术重点实验室;
【基金】：国家高技术研究发展计划(2012AA121103)
【分类号】：TP732

【相似文献】