CAJ | 학술논문

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。
당전기우단폭무천도상복원산법적연구도시재광조충족적정형하（백천），저충정형하적광조주요래자천공광。연이，대우재광조주요래자인조광（야간）적무천도상복원산법적연구굴지가수。인차제출료신적기우인조광조하적무천성상모형，병동시고필료불균균적배경광화대도상적광조보상。검측출광원위치，근거광원적발광특성구출배경광화광조보상계수，이용제출적신모형구출청석적도상，사용기우변연안색항상성（EBCC）래대안색진행교정。실험결과표명，해산법우우기타산법。
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 illumina-tion 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.