基于混合像元分解的偏振高光谱图像去雾方法

Fog Removal Method for Polarized Hyperspectral Images Based on Mixed Pixel Decomposition

常规的偏振高光谱图像去雾方法在应用过程中存在端元提取精度偏低的问题,使得偏振高光谱图像的去雾效果不佳,因此,基于混合像元分解设计了新的偏振高光谱图像去雾方法。分析了偏振高光谱图像散射偏振特性,降低了线性混合误差,基于混合像元分解提取了偏振高光谱图像端元,获取了最优约束。通过高光谱图像去雾算法提高了端元的提取精度,实现了偏振高光谱图像的有效去雾。实验结果表明,所设计的偏振高光谱图像去雾方法的端元提取精度较高,去雾效果较好,具有较好的应用性能,为光谱图像处理提供了一定的技术参考。

The conventional polarization hyperspectral image defogging method has the problem of low end element extraction accuracy in the application process,and the defogging effect of polarization hyperspectral image is poor.Therefore,this paper designs a new polarization hyperspectral image defogging method based on mixed pixel decomposition.The scattering polarization characteristics of polarized hyperspectral images are analyzed to reduce the linear mixing error.Based on the mixed pixel decomposition,the end elements of polarized hyperspectral images are extracted to obtain the optimal constraints.The precision of end element extraction is improved by using the defogging algorithm of hyperspectral images,and the effective defogging of polarized hyperspectral images is realized.The experimental results show that the designed polarization hyperspectral image defogging method has higher end element extraction accuracy,better defogging effect and good application performance,which provides the certain technical support for spectral image processing.