煤矿智能视频监控中雾尘图像的清晰化研究

Clearing research on fog and dust images in coalmine intelligent video surveillance

煤矿智能视频监控中常常碰到许多雾尘图像且伴有各种随机噪声,对应的图像降质严重影响了后续视频图像处理工作,因此提出一种基于暗原色先验与双边滤波器的去雾除尘和同步去噪算法。结合已有的大气散射物理模型,推导建立煤矿雾尘图像退化模型。考虑煤矿雾尘图像的特点,设计基于暗原色先验知识的大气光、粗略透射率估计的方法与步骤。分析粗略透射率图的优化要求以及双边滤波器的特性,引入联合双边滤波器快速获得精细透射率图。依据图像退化模型构建正则化目标函数,求取转换图像并进行高斯双边滤波,获得复原图像并同步实现噪声的有效去除。实验结果验证了算法的有效性,与已有去雾算法相比计算效率有较大提高,且复原质量良好适合于煤矿智能视频监控环境。

There are many fog and dust images with much random noise in coalmine intelligent video surveillance. Therefore, the image degradation has seriously affected the subsequent video image processing. In this paper, an algo- rithm based on dark channel prior and bilateral filter was proposed which can realize fog and noise simultaneous｜y re- moving. By combining the atmospheric scattering model, the degradation model of fog and dust images was established. Considering the characteristics of fog images, the methods and procedures for estimating the air light and rough trans- mittance were designed using the dark channel prior. By analyzing the optimization requirement of the rough transmission map and the bilateral filter characteristic, a joint bilateral filter was introduced for quickly obtaining the fine transmission map. The regularization objective function was constructed on the image degradation model. By solving a converted image and Gaussian bilateral filtering the image, fog and dust removal and simultaneously denoising were real- ized. Experimental results verify that the proposed algorithm is effective which computational efficiency is greatly improved compared with various restoration algorithms. Because of good restoring quality the algorithm is suitable for the environment of coalmine intelligent video surveillance.