计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (20): 200-203.

• 图形图像处理 • 上一篇    下一篇

基于透射率归一化的单幅图像去雾算法

刘恩雨,朱海涅,谢建磊   

  1. 武汉大学 电子信息学院,武汉 430079
  • 出版日期:2016-10-15 发布日期:2016-10-14

Single imagine defogging based on transmission normalization

LIU Enyu, ZHU Hainie, XIE Jianlei   

  1. School of Electronic Information, Wuhan University, Wuhan 430079, China
  • Online:2016-10-15 Published:2016-10-14

摘要: 在有雾天气条件下拍摄的图像,由于光线在传播人过程中受到空气中悬浮颗粒的散射,导致图像内容模糊不清,颜色偏灰白色。为了恢复出清晰的图像,根据大气散射物理模型,提出一种单幅图像去雾算法。首先,该算法从大气散射模型出发,对大气光模型进行变形化简,得到新的去雾模型。然后对大气光值估计,并利用在局部统计上不相关性恢复透射率。最后,对透射率进行归一化处理,带入模型,得到恢复图像。实验结果表明,该算法能够很好地恢复图像的对比度和保持图像的真实颜色。

关键词: 图像去雾, 大气光值, 透射率, 归一化

Abstract: Because of light scattered by the suspended particles in the atmosphere, photographs taken in the foggy day look gray are lack of visibility. In order to unveil the clear image’s structures and colors, it proposes a new algorithm based on the atmosphere scattering model using a single image. Based on atmosphere scattering model, it simplifies the atmospheric scattering model from the perspective of fog formation and obtains new dehazing model. Then, it estimates the atmosphere light value, and it estimates the transmission by using the uncorrelation principle. Finally, it uses normalization method to adjust the transmission value, and takes into the new simplified model to get the dehazing image. In the experiments, the algorithm has a good effect on restoring the clear image’s contents and preserving the image’s true colors.

Key words: image dehazing, atmosphere light value, transmission, normalization