多光谱遥感技术在研究地表生物量、 气候变迁和自然灾害预报等方面有重要应用, 遥感器的光谱敏感度与遥感数据的定量化存在相关性。 遥感器在轨运行期间, 受空间辐射、 温度剧变和化学分子污染或宇宙尘埃等因素的影响, 光学系统光谱敏感度产生退化, 导致测量值与真值出现偏差, 影响影像产品的精度。 在已有工作的基础上, 提出了一种简化的基于人工光谱检测参照的光谱响应特性在轨评估方法, 通过建立光谱响应函数的退化模型, 结合光谱靶标反射率的测量结果与影像反演结果, 实现遥感器光谱响应特性的在轨评估。 经仿真实验与在轨实验验证, 该方法可以有效实现对光谱响应函数波长漂移和带宽缩放变化的检测, 并有利于提高遥感器长时间序列数据产品精度和多种遥感器数据产品的综合利用水平。

The multispectral remote sensing technology has been a primary means in the research of biomass monitoring, climate change, disaster prediction and etc. The spectral sensitivity is essential in the quantitative analysis of remote sensing data. When the sensor is running in the space, it will be influenced by cosmic radiation, severe change of temperature, chemical molecular contamination, cosmic dust and etc. As a result, the spectral sensitivity will degrade by time, which has great implication on the accuracy and consistency of the physical measurements. This paper presents a characterization method of the degradation based on man-made spectral targets. Firstly, a degradation model is established in the paper. Then, combined with equivalent reflectance of spectral targets measured and inverted from image, the degradation characterization can be achieved. The simulation and on orbit experiment results showed that, using the proposed method, the change of center wavelength and band width can be monotored. The method proposed in the paper has great significance for improving the accuracy of long time series remote sensing data product and comprehensive utilization level of multi sensor data products.