摘要
为改善雾天拍摄图像的清晰度,提出一种实用的图像去雾方法。先利用暗原色先验算法去除图像雾干扰,再运用插值抠图算法进一步改善图像去雾效果,最后利用图像的区间值信息对其进行放大,以便获得满意的视觉效果。实验结果表明,该去雾方法不仅可直接估计雾的浓度,并且可得到去雾后的高清晰图像。
In order to improve the sharpness degree of images captured in the condition of foggy weather,an useful defogging method for image is proposed.First,a dark channel prior algorithm is utilized to remove the fog interference from a single input image.Then it uses a soft matting interpolation algorithm to further improve the image effect.Finally,the interval information of image is used to magnify the images in order to obtain satisfactory visual effect of the image.The experimental results show that the defogging method proposed can not only directly estimate the concentration of the fog,but also recovery remove fog interfere with a clear image of high-quality.
出处
《西安邮电学院学报》
2012年第6期27-32,共6页
Journal of Xi'an Institute of Posts and Telecommunications
基金
陕西省教育厅自然科学专项基金资助项目(2010JK816)
关键词
图像去雾
暗原色先验
插值抠图算法
清晰图像
image dehazing
dark channel prior
interpolation matting algorithm
haze-free image
