基于大气散射模型的偏振图像去雾算法

Polarization dehazing algorithm based on atmosphere scattering model

雾是1种常见的天气现象,由于大气中的水汽等对太阳光的散射作用会造成拍摄图像的对比度降低。在大气散射物理模型的基础上,提出1种可以轻松去除图像雾霾的方法。该算法首先利用偏振成像系统获取最好状态下和最差状态下的2幅图像,然后通过边缘检测和最佳正态分布搜索算法对天空区域进行分割,从而估算大气光强度和大气传递系数,最终实现图像去雾。在Matlab平台上进行仿真的结果表明了该算法不仅可以有效改善雾天下图像的退化现象,而且可以提高图像的清晰度。

Haze which is caused by scattering of sunlight by water vapor and other scattering sources in atmosphere,can significantly reduce the contrast of images.Present an approach to easily remove the effects of haze from images based on the model of atmosphere scattering.The algorithm makes use of polarization imaging detection system to obtain two images under the best and the worst polarization states first.Then uses edge detection algorithm and the best normal distribution search algorithm to divide the sky,and the atmosphere light intensity and atmosphere transfer coefficient are estimated.Simulation results in the Matlab platform demonstrate that this algorithm can efficiently improve the degradation of image and enhance the clearness of image.