针对高光谱图像溢油检测过程中云背景的干扰问题，提出一种高光谱场景云背景抑制方法.首先，分析了海面溢油的光谱反射率特性，根据海面溢油中C-H键的光谱特征，介绍了基于伪彩图像的溢油检测方法.其次，比较了高光谱场景中海水、溢油和云背景的辐射特性，依据云背景辐射特性的特点，设计了一种新的云背景辐射光谱特征提取模型，在此基础上，进一步考虑云、海面与溢油的差异，选取云辐射最大的波段图像，结合云背景辐射光谱特征生成云背景抑制图.最后，将云背景抑制图与溢油伪彩图像相乘，得到云背景抑制结果.将本文提出的方法应用于实际墨西哥湾Airborne Visible InfraRed Imaging Spectrometer高光谱遥感影像的海面溢油的检测，结果表明本文提出的方法能够在不影响溢油检测条件下，有效地消除高光谱溢油检测过程中云背景的影响.

Hyperspectral sensors can acquire the radiance of a scene in a high resolution both in spatial and spectral dimensions, which produces good benefits for oil spill detection. Because the cloud background in hyperspectral scene severely interfere oil spill detection result, a new method is presented to reduce the cloud background in hyperspectal scenes. Firstly, the spectral reflectance characteristics of oil spill are analyzed. According to the spectral characteristics of the C-H bond of oil spill, the false color generation based oil spill detection method is introduced. Secondly, the radiance characteristics of seawater, oil spill and cloud is compared. Based on the radiation characteristics of cloud, a new model is build to extract the radiance features of cloud. On this basis, the difference of cloud and seawater, and oil spill are considered. And the band image which has maximum cloud radiance is selected. The cloud background suppression map is then generated by using the band image and the radiance feature of the cloud background. Finally, the background suppression result is obtained by multiplying the false color image by background suppression map. The proposed method is applied to the real Airborne Visible InfraRed Imaging Spectrometer hyperspectral image captured during the Deepwater Horizon oil spill in the Gulf of Mexico for oil spill detection. The results show that the proposed method can effectively suppress the cloud background for oil spill detection from hyperspectral data, and does not affect the oil spill detection performance.