基于暗原色先验理论在图像的去雾方面有非常好的效果，提出了一种将暗原色先验理论用于大气能见度测量的方法，克服了气象能见度仪成本昂贵且难以大范围架设的缺点。对拍摄雾霾图像的不同区域进行亮度分析以便选择合适的目标区域进行透射率估计，通过优化去雾系数并使用引导滤波细化透射率获取较为准确的透射率，从而求得大气消光系数及大气能见度。将大量雾霾图像处理的结果与前向散射仪(CJY-1G）测量的数据进行比较，二者基本一致，本方法与CJY-1G测量结果的误差在±15%之内，符合能见度观测的应用要求。

Dark channel prior has a good effect on haze removing. A visibility measurement method based on dark channel prior is proposed, which can overcome the disadvantages of high cost and difficulty in building up of the meteorological visibility meter. The proper target region is selected by analyzing different regions of the image so as to estimate transmittance. A more accurate transmittance is obtained by haze removal parameter optimization and guided filtering refinement, and then the atmospheric extinction coefficient and visibility are inferred. Comparing the results of dark channel prior with the data measured by forward scattering meter (CJY-1G), the error is within ±15%, which meets the application requirement of World Meteorological Orgnization.