摘要： 针对雾天图像景物容易退化的问题，提出一种基于暗原色和插值直方图匹配的雾天图像自适应增强算法。该算法基于暗原色先验信息寻找并分割出图像中的天空区域，利用插值直方图匹配方法对非天空区域进行自适应增强处理，采用主成分分析方法将处理前后的图像信息进行融合。实验结果表明，该算法不仅能避免天空噪声对图像的干扰，而且能有效提高图像对比度，获得满意的图像视觉效果。

关键词: 去雾, 暗原色, 直方图匹配, 图像增强

Abstract: According to the degraded effect of images obtained in foggy weather, a new enhancement algorithm based on dark channel prior and local interpolating histogram matching is proposed. The sky region is separated from the degraded image using dark channel prior. The local interpolating histogram matching algorithm is adopted to deal with the non-sky region of the image. The disposals of after and before images are fused by principal component analysis. Experimental results show that the algorithm can not only efficiently avoid interfere with sky regions, but also enhance the contrast of degraded images and obtain satisfying visual effects.

Key words: defogging, dark channel prior, histogram matching, image enhancement

中图分类号: 

• N945