基于改进饱和线先验的遥感图像去雾算法

Remote Sensing Image Dehazing Algorithm Based on Improved Saturated Line Prior

针对在雾霾天气下获得的遥感图像产生清晰度降低,对比度和色彩保真度下降的问题,考虑到遥感图像成像较宽、信息量较大的特点,本文提出一种基于改进饱和线先验的遥感图像去雾算法。该算法首先对初始图像进行预处理,其次基于饱和线先验理论对含雾图像构建饱和线以估计初始透射率,之后引入补偿因子与梯度域导向滤波器对透射率进行优化,提升了算法的鲁棒性,最后根据大气散射模型复原出清晰遥感图像。数值实验结果表明,本文算法对多种场景下的含雾遥感图像都取得了良好的效果。Aiming at the problem of reduced clarity, contrast and color fidelity of remote sensing images obtained under hazy weather, considering the characteristics of wide imaging and large amount of information in remote sensing images, we propose the remote sensing image dehazing algorithm based on the improved saturated line prior. We first preprocess the initial image, then construct saturated lines based on the saturated line prior theory for hazy image to estimate the initial transmittance, then introduce a compensation factor and the gradient-domain oriented filter to optimize the overall transmittance, which improve the robustness of the proposed algorithm, and finally recover a clear remote sensing image based on the atmospheric scattering model. Numerical experimental results show that the proposed algorithm achieves good results for hazy remote sensing images in a variety of scenes.