基于颜色衰减的自适应去雾算法

雾霾使室外拍摄的图像、视频画质退化严重，给室外安防和交通监控等系统的正常运行带来困难。去雾算法旨在恢复图像质量，增强图像对比度和清晰度。本文提出了一种结合大气散射模型与颜色衰减先验的去雾复原模型，并以新增可见边比为评价标准，给出了模型参数的自适应求取方法，并采用引导滤波对透射率进行优化，从而较好地恢复出无雾图像。对有雾图像分别采用本文方法和三种现有去雾算法进行对比实验，从实验结果看，基于颜色衰减的自适应去雾算法可使图像清晰度、对比度得到较大的提高，与其他算法相比，在实时性和清晰度方面有一定优势。

An Adaptive defogging Algorithm Based on Color Attenuation

Abstract: The haze makes the outdoor images and videos quality degraded seriously, which brings big difficulties to the normal operation of security and traffic monitoring system. Defogging algorithm is designed to to restore the image quality, enhance the image contrast and clarity. In this paper,a model of defogging restoration model is proposed based on the atmospheric scattering model and the color attenuation prior. Furthermore, by introducing additional visible edge ratio as the evaluation criterion, an adaptive method for obtaining the model parameters is proposed, then a set of adaptive algorithm for defogging is formed, and then the transmission map is optimized by the guided filter to recover the final haze-free images. Experiments have been taken by applying the presented method and three existing defogging algorithms to two foggy images. the experimental results show that the adaptive defogging algorithm in this paper can make the image sharpness and contrast improved greatly, and has advantage in real-time and clarity comparing with the other three algorithms.