Abstract: Firstly, the global atmospheric light was calculated using the Dark Channel Prior (DCP). Secondly, the global atmospheric light and the atmospheric veil estimated were used to compute the media transfer rate. Finally, the recovered image could be obtained using the atmospheric attenuation model. This algorithm combined both the priorities of the accurate estimation of global atmospheric light by DCP and the fast calculated procedure of the atmospheric veil. The experimental results demonstrate that the proposed method not only achieves good visual recovery results, but also shortens the time to remove images haze.

Key words: Haze removal, Dark channel prior, Airlight, Atmospheric attenuation model, Media transfer rate

摘要： 运用黑色通道先验知识(DCP)计算出整体大气光,然后通过该整体大气光和估计的大气散射光求解出介质传输率,最后由大气衰减模型得到复原结果。该算法集DCP估算整体大气光较为精确的优点和大气散射光计算快速的优点于一体。实验结果表明,该算法不仅能在视觉效果上取得较好的复原结果,而且使得去除雾霾的时间大大缩短。

关键词: 雾霾去除, 黑色通道先验知识, 大气光, 大气衰减模型, 介质传输率