结合天空识别和暗通道原理的图像去雾-中国图象图形学报.DOC

结合天空识别的图像去雾 摘 要：目的 目前较为流行的去雾算法对天空区域的处理效果不佳，容易造成方块效应以及色彩严重失真。针对该问题，提出了一种基于天空识别和暗通道原理的单幅图像去雾方法。方法 由于暗通道原理在天空区域失效，故提出天空识别算法，将雾图分为天空与非天空部分，对其分别估计透射率图，通过大气散射模型得到复原图像；其次，针对利用暗通道原理得到的去雾图像普遍偏暗的问题，对其进行色彩重映射，以增加图像亮度，提升图像视觉效果。结果 大量实验结果表明，本算法复原的图像清晰自然，尤其是天空区域平滑明亮，取得了很好的去雾效果。结论 基于天空识别，提出了一种新颖的单幅图像去雾算法。与He Kaiming以及Tarel的算法相比，去雾后图像整体效果更佳。 关键词：天空识别；暗通道原理；图像去雾；色彩重映射 Li Jiayuan1, Hu Qingwu1, Ai Mingyao1,2, Yan Jun1 1.School of Remote Sensing and Information Engineering, Wuhan, 430079. 2. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan, 430079. Abstract: Objective Images captured in foggy weather are often degraded due to atmospheric absorption and scattering, and haze removal is highly desired in both image processing and computer vision applications. First, removing haze can significantly increase the visibility of the scene. Second, most image processing and computer vision algorithms, usually assume that the input image is the scene radiance. Therefore, many methods have been proposed, while sky region processed by most of these algorithms tends to be degraded by block noise and serious color distortion. Concerning this issue, an improved single image haze removal method based on sky region detection and dark channel prior is proposed. Method Our method consists of three major stages: sky region detection, haze removal and tone mapping. In the first stage, we observe that sky is usually a large and smooth region with high intensity. Based on these characteristics, an effective algorithm is designed to divide the input image into sky and non-sky region. In the next stage, dark channel prior is used to estimate the transmission maps of the two regions and guided filter is applied to refine these maps, then, the haze-free image can be recovered by atmosphere scattering model. The final stage uses a simple tone mapping algorithm to increase the image brightness which leads to good visual effects. Result In He’s haze removal algorit