中文摘要：

随着计算机视觉系统的发展及其在军事、交通以及安全监控等领域的发展，图像去雾已成为计算机视觉的重要研究方向．在雾、霾之类的恶劣天气下采集的图像会由于大气散射的作用而被严重降质，使图像颜色偏灰白色，对比度降低，物体特征难以辨认，不仅使视觉效果变差，图像观赏性降低，还会影响图像后期的处理，更会影响各类依赖于光学成像仪器的系统工作，如卫星遥感系统、航拍系统、室外监控和目标识别系统等．因此，需要图像去雾技术来增强或修复，以改善视觉效果和方便后期处理．本文归纳总结了两大类图像去雾方法：基于图像增强和基于物理模型的方法，深入探讨了其中的典型算法和研究成果，并对这些算法的测试结果进行了定性和定量的分析比较，最后总结了图像去雾技术目前的研究状况和未来的发展方向．

英文摘要：

With the development of computer vision system and the increasing demand in military, transportation and surveillance applications, image dehazing has been an important researching direction in computer vision. Images acquired in bad weather, such as haze and fog, are seriously degraded by the scatting of the atmosphere, which makes the image color gray, reduces the contrast and makes the object features difficult to identify. The bad weather not only leads to the variation of the visual effect of the image, but also cause the disadvantage of the post processing to the image, as well as inconvenience of all kinds of instruments which rely on optical imaging, such as satellite remote sensing system, aerial photo system, outdoor monitoring system and object identification system. That is the reason why the image need enhancement and restoration for the improvement of the visual effects and convenience of post processing. This paper sums up two kinds of image dehazing methods, which are the methods based on image enhancement and based on the physics model. After that, some algorithms and research results are presented, followed by quantitative and qualitative evaluations of these techniques. Finally, the research progress is summarized and future research directions are suggested.