基于饱和度的多尺度雾天降质图像质量增强算法研究

传统图像质量增强算法只适于薄雾状态下的降质图像,对浓雾状态下图像质量的增强效果较差。为此,提出一种新的基于饱和度的多尺度雾天降质图像质量增强算法,通过混合灰度转换函数子带分解多尺度Retinex算法挑选高、中、低三个尺度,结合雾天降质图像整体阴影区域和高光部分的细节,依次完成对雾天降质图像各个频段的质量增强,获取各个频段的质量增强结果。把获取结果与原图像共同视为一个图像集合,通过图像融合技术完成对所有图像的权重图分配操作,提高增强后图像质量。实验结果表明,所提算法能够有效增强雾天降质图像质量,主观客观评价结果均较优。

Research on image quality enhancement algorithm for multi scale fog degraded image based on saturation

The traditional image quality enhancement algorithm is only suitable for the degraded image in the fog condition, and the image quality enhancement effect is poor. For this, put forward a new kind of fog based on multi-scale saturation enhancement algorithm for image quality quality of heaven, through the mixed gray transfer function subband decomposition of multi-scale Retinex algorithm to extract high, medium and low three scale, combined with fog degraded image overall shadow region and the high part of the details, in order to complete the quality degradation the image of each band fog enhancement, quality of each frequency enhancement results. The results are compared with the original image as a set of images, and the image fusion technology is used to assign the weights of all images. The experimental results show that the proposed algorithm can effectively improve the quality of fog degraded images, and the subjective and objective evaluation results are better.