基于暗原色先验与WLS的图像去雾算法

Image Dehazing Method Based on Dark Channel Prior and WLS

雾霾天气条件下图像采集因降质严重,导致后期图像处理复杂性提高。为此,提出一种采用暗原色先验理论与WLS滤波相结合去雾的改进算法。采用WLS滤波代替传统的软抠图法以修复大气透射率,针对去雾后图像较实际暗沉的问题,提出一种新的自适应图像增强方法,通过对去雾后的图像自适应非线性叠加,实现了图像增强的效果。实验结果表明,相比于其它传统算法,该算法能够在保持图像边缘细节的同时,提高图像的色彩质量,有效消除白色晕块,且计算复杂度低,图像还原逼真。

Images captured in fog and haze weather conditions suffer from serious degradation,which will enhance the complexity of images post-processing.To solve the problem,an improved algorithm which combines the dark channel prior and WLS filter is proposed.First,WLS filter is used to repair atmospheric transmission instead of the traditional soft matting method.Then,a new adaptive image enhancement method is proposed,which aims at the problem that the restored image is duller than the actual one.It achieves the effect of image enhancement by the adaptive non-linear superposition of the dehazing image.The experimental results show that compared with other traditional algorithms,the proposed algorithm can improve the color quality of the image while maintaining the details of the image edge.It effectively eliminates the white halo.In addition,its computational complexity is lower and the restored images are approximates the originals.