基于暗原色先验的图像快速去雾

针对雾天图像饱和度和对比度较低,提出一种基于暗原色先验和透射率权值的改进算法。估计大气光的取值范围,根据该范围确定大气光值,通过判定因子区分出暗原色失效区域;对该区域透射率进行基于透射率权值因子的调整;采用快速引导滤波细化透射率,并结合大气散射模型恢复图像。该算法能在去雾过程中对可调参数进行自适应选取。实验结果表明,该算法效率较高,且复原的图像细节丰富、颜色自然,视觉效果提升的较为明显。

IMAGE FAST DEHAZING BASED ON DARK CHANNEL PRIOR

In view of the low saturation and contrast of hazy images,this paper proposes an improved algorithm based on dark channel prior and transmittance weight.The range of atmospheric light was estimated and the value of atmospheric light was determined according to this range.Then,the transmittance in this region was adjusted based on the weighting factor of transmittance.The fast guided filtering was used to refine the transmittance,and the image was restored by combining with the atmospheric scattering model.The proposed algorithm could select the adjustable parameters adaptively in the process of image dehazing.Experimental results show that the algorithm is efficient,and the image restored by the algorithm is rich in details,natural in color,and the visual effect is improved obviously.