自适应高斯融合双通道的去雾算法

Adaptive Gaussian Fusion Double Channel Dehazing Algorithm

针对传统图像去雾算法存在的对比度下降和颜色偏移等问题,提出一种结合高斯融合的自适应双通道雾霾图像复原算法。首先,考虑到大气光应小于有雾图像最大值,且大于有雾图像最小值,根据亮度控制因子自适应控制的方式得到融合中通道,并获得中通道下的局部大气光;其次,提出双通道线性传输,即用最大值通道辅助完成线性传输,再用高斯函数加权融合的方法实现清晰图像最优通道估计,从而得到最优透射率;最后,结合复原模型恢复清晰图像。实验表明,所提方法有效解决了图像对比度下降与颜色偏移等问题,去雾效果良好、亮度适宜且颜色保真度更高。另外,该方法在定量指标上同样具有优越的表现。

Aiming at the problems of incomplete dehazing and color distortion in the traditional image dehazing algorithm,a dehazing algorithm based on double channel with adaptive Gaussian fusion is proposed.First,considering that the atmospheric light should be less than the maximum value of the foggy image and greater than the minimum value of the foggy image,the fusion middle channel is obtained according to the method of adaptive control of the brightness control factor,and the local atmospheric light under the middle channel is obtained.Secondly,the double channel linear transmission is proposed,that is,the maximum channel is used to assist in the linear transformation,and then the Gaussian function weighted fusion method is used to realize the optimal channel estimation of the clear image,thereby obtaining the optimal transmission.Finally,the clear image is restored by combining the restoration model.Experiments show that the proposed method effectively solves the problems of image contrast decreases and color shift,with good dehazing effect,suitable brightness and higher color fidelity.In addition,the method also has excellent performance in quantitative indicators.