人造光照下的单幅雾天图像复原  

Single image de-haze under illumination from artificial light sources

作　　者：鲁丹[1] 方帅[1,2] 邵堃[1] 

机构地区：[1]合肥工业大学计算机与信息学院,合肥230009 [2]中国科学院安徽光学精密机械研究所,合肥230031

出　　处：《计算机工程与应用》2015年第22期187-192,共6页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61175033);中央高校基本科研业务费专项资金(No.WK2100100009,No.2012HGCX0001)

摘　　要：当前基于单幅雾天图像复原算法的研究都是在光照充足的情形下(白天),这种情形下的光照主要来自天空光。然而,对于在光照主要来自人造光(夜间)的雾天图像复原算法的研究屈指可数。因此提出了新的基于人造光照下的雾天成像模型,并同时考虑了不均匀的环境光和对图像的光照补偿。检测出光源位置,根据光源的发光特性求出环境光和光照补偿系数,利用提出的新模型求出清晰的图像,使用基于边缘颜色恒常性(EBCC)来对颜色进行校正。实验结果表明,该算法优于其他算法。Current researches on single image de-hazing always premise a sufficient illumination, which refers to skylight in the daytime. However in condition that illumination mainly comes from artificial light sources, the current algorithm will not work well and related researches are rare. This paper presents a novel image de-hazing algorithm under illumination from artificial light sources, which considers non-uniform illumination and light compensation. A detecting algorithm is proposed to confirm the position of artificial light sources; atmospheric light and compensatory coefficient are obtained according to artificial light characteristic; after that, the haze-free image is effectively recovered based on a newly proposed imaging model; edge based color constancy is used to correct color. Experimental results show that this algorithm can get good performance under non-uniform illumination.

关 键 词：去雾 人造光 颜色恒常性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]