为了提高雾天视频的可用性, 提出了一种基于暗通道先验和区间估计的视频去雾方法。利用暗通道先验知识, 采用区间估计的方式, 运用图像融合的思想, 通过背景图像求取大气光值和介质传输率, 并应用于视频的所有帧以去除雾气。与几种典型的视频去雾方法相比, 具有较快的运算速度, 能有效地避免去雾视频中出现颜色跳变的问题。以暗通道先验理论为基础, 采用区间估计的方式对大气光值和介质传输率进行估计, 从而能有效地提高去雾视频的整体亮度、清晰度和对比度, 同时获得较好的图像颜色。

To improve the usability of foggy video, a video defogging method based on dark channel prior and interval estimation is proposed. Through the dark channel prior knowledge, interval estimation and image fusion are adopted to estimate the value of global atmospheric light and medium transmission from the background image, which are applied to a series of video frames to eliminate the fog. Compared to some state-of-the-art methods, the proposed method can achieve high processing speed and effectively avoid color variations in restored video. The value of global atmospheric light and medium transmission using the interval estimation are estimated by the method, which is based on dark channel prior. The method can effectively improve the overall brightness, visibility and contrast of restored video, and obtain good color effect.