图像和视频的快速去雾算法研究

雾、霾等恶劣天气会导致室外图像能见度和对比度降低。虽然可以通过增强有雾图像的对比度得到清晰的图像,但对比度的过度增强可能会截断像素值,造成信息丢失。因此,本文基于信息丢失问题提出了一种快速、优化的去雾算法。通过最小化信息丢失,使输出图像不仅能保留较多的细节,且具有较高的对比度。此外,通过将RGB颜色空间转换为YUV颜色空间,仅对亮度分量Y进行处理,提高了算法的运算速度。算法的对比实验结果表明,本文的算法不仅去雾效果明显,而且运算速度快,完全能满足视频去雾的实时性要求。

Fast Haze Removal for Real-time Image and Video

Outdoor image often lose visibility and contrast due to the presence of haze. To effectively enhance the degraded haze image, we can get clear image by enhancing haze image's contrast. However, the excessive enhancement of the degraded contrast may truncate pixel values and cause information loss. Therefore, we proposed a fast and optimized dehazing algorithm based on information loss. By minimizing the information loss, the output image not only has high contrast but also has very little information loss. Moreover, we convert RGB color space into YUV color space, and only process the luminance component. This greatly improves the computational speed, thus suitable for processing video sequence. The contrast experimental results show that the proposed algorithm can effectively remove haze and the computing speed is fast, which can fully meet the real-time requirement for the removal of video haze.