基于大气传输函数的单幅图像去雾算法

Single Image Dehazing Algorithm Based on Atmosphere Transfer Function

针对雾霾退化场景图像的反照率恢复问题,采用大气传输函数对图像雾霾退化过程进行描述。采用雾霾退化条件下的大气传输函数曲线,通过对模型的分析,结合单色大气退化模型的相关先验知识,构建极小化能量泛函;通过切分Bregman迭代算法进行求解,实现对大气传输函数中各系数的求解,进而实现对场景反照率的恢复。通过仿真实验表明,算法能够较好地实现对各类雾霾退化场景的恢复,有效提高图像视见度;且算法具有较高的处理效率与鲁棒性。

To solve the issue of image degradation caused by fog and haze, the atmosphere transfer function is induced to depict the process of degradation. The curve modeling condition is combined with the atmospheric scat- tering model. Consulting prior knowledge of atmospheric scattering model, the energy functional is utilized to obtain the coefficients of the function with the split Bregman. And then the scene albedo is recovered. The experiment shows that the algorithm is valid and robust for all kinds of scene and enhances the visibility with high efficiency.