基于暗通道先验的图像去雾优化算法

暗通道先验算法虽然在单幅图像去雾方面取得了一定的效果,但是该算法运行时间较长,另外对环境光的计算不太准确,不适用于天空区域,会导致复原图像色彩失真、亮度偏暗。针对这些缺陷,本文提出一种改进的White Patch Retinex算法,对原有图像去雾算法进行优化。首先,通过改进的White Patch Retinex算法计算出环境光。其次通过暗通道先验算法获得透射率。最后根据得到的环境光和透射率,求解大气散射模型,从而得到去雾后的图像。实验结果表明,该算法不仅运行时间较短,对分辨率为600×800的图像处理时间平均为5 s左右,且能解决天空区域失真问题,去雾后的图像具有较高的亮度和对比度。

Optimized dehazing method based on dark channel prior

Although the dark channel prior algorithm is effective to some extent for single image dehazing, the algorithm has some disadvantages of running a long time. Besides, the calculation of environmental light is not very accurate, and not suitable for the sky area, which can lead to the color distortion and dim brightness for the restoration image. Aiming at theses hortcomings, this paper presents an optimization algorithm. Firstly, the environmental light is calculated by the improved White Patch Retinex algorithm. Secondly, the transmittance is obtained based on the dark channel prior. Finally, according to the atmospheric scattering model, the final haze-free images are obtained. Experimental results show that the proposed algorithm has a shorter running time. The average processing time of the image (600×800 pixel) is about 5 s.The problem of color distortion for the sky area can be well solved.After defogging, the image has high brightness and contrast.