Nighttime Fog Removal Using the Dark Point Light Source Model
-
摘要: 针对夜间雾、霾场景下的去雾图像存在颜色失真、纹理损失、亮度低等缺陷，本文提出了一种采用暗态点光源模型的夜间去雾算法，通过构建夜间雾、霾场景的暗态点光源模型，利用联合双边滤波、限制对比度自适应直方图均衡化等算法对降质图像进行处理，结合大气散射模型得到去雾图像。实验结果表明，该算法的处理速度快、夜间去雾效果较好，较对比算法在对比度、平均梯度以及信息熵上均有一定程度地改善，有效解决了去雾图像的颜色失真、纹理损失、亮度低等缺陷。
-
关键词:
- 夜间去雾 /
- 暗态点光源模型 /
- 联合双边滤波 /
- 限制对比度自适应直方图均衡化 /
- 大气散射模型
Abstract: To address image distortion, texture loss, and low brightness in nighttime fog scenes, this paper proposes a nighttime defogging algorithm based on a dark point light source model. The dark point light source model was first constructed and the degraded image was processed by an algorithm that utilizes both bilateral filtering and limited contrast adaptive histogram equalization. Then, the defogging image was obtained by combining with the atmospheric scattering model. The experimental results show that this algorithm has a fast processing speed, a better effect of nighttime fogging, and a certain degree of improvement in terms of contrast, average gradient, and information entropy when compared with the contrast algorithm. This model can therefore effectively address image distortion, texture loss, and low brightness of fogging images. -
-
Index Algorithm Fig. 1 Fig. 2 Fig. 3 Fig. 4 IE HE[7] 7.3539 6.9861 6.9447 6.5511 Zhang[4] 7.0074 6.6843 6.8380 6.4599 Li[5] 6.2890 5.8723 6.4657 5.4543 Proposed 7.5694 6.7746 7.0220 6.6502 AG HE[7] 0.0646 0.0693 0.0521 0.0415 Zhang[4] 0.0491 0.0519 0.0471 0.0311 Li[5] 0.0272 0.0276 0.0288 0.0161 Proposed 0.0653 0.0707 0.0535 0.0457 Contrast HE[7] 0.1745 0.1709 0.1346 0.1059 Zhang[4] 0.1565 0.1622 0.1528 0.1217 Li[5] 0.1129 0.0913 0.1183 0.0875 Proposed 0.1823 0.1754 0.1562 0.1238 
下载: 导出CSV
-
[1] 余顺园, 朱虹. 夜间有雾图像的光照模型构建及去雾[J]. 光学精密工程, 2017, 25(3): 729-734. https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201703025.htm YU S Y, ZHU H, Illumination model construction and defogging of night foggy image[J]. Optics And Precision Engineering, 2017, 25(3): 729-734. https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201703025.htm
[2] 郭璠, 邹北骥, 唐琎. 基于多光源模型的夜晚雾天图像去雾算法[J]. 电子学报, 2017, 45(9): 2127-2134. DOI: 10.3969/j.issn.0372-2112.2017.09.011 GUO P, ZOU B J, TANG J. Defogging algorithm of night fog image based on multi light source model[J]. Acta Electronica Sinica, 2017, 45(9): 2127-2134. DOI: 10.3969/j.issn.0372-2112.2017.09.011
[3] 左健宏, 蔺素珍, 禄晓飞, 等. 基于雾线暗原色先验的红外图像去雾算法[J]. 红外技术, 2020, 42(6): 552-558. http://hwjs.nvir.cn/article/id/hwjs202006007 ZUO J H, LIN S Z, LU X F, et al. Infrared image defogging algorithm based on fog line dark primary color prior[J]. Infrared Technology, 2020, 42(6): 552-558. http://hwjs.nvir.cn/article/id/hwjs202006007
[4] ZHANG J, CAO Y, FANG S, et al. Fast haze removal for nighttime image using maximum reflectance prior[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7418-7426.
[5] LI Y, TAN R T, Brown M S. Nighttime haze removal with glow and multiple light colors[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015: 226-234.
[6] YU T, SONG K, MIAO P, et al. Nighttime Single Image Dehazing via Pixel-Wise Alpha Blending[J]. IEEE Access, 2019(7): 114619-114630. http://ieeexplore.ieee.org/document/8805086
[7] HE K, SUN J, TANG X. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 33(12): 2341-2353.
[8] 刘志成, 王殿伟, 刘颖, 等. 基于二维伽马函数的光照不均匀图像自适应校正算法[J]. 北京理工大学学报, 2016, 36(2): 191-196, 214. https://www.cnki.com.cn/Article/CJFDTOTAL-BJLG201602016.htm LIU Z C, WANG D W, LIU Y, et al. Adaptive correction algorithm for uneven illumination image based on two dimensional gamma function[J]. Journal of Beijing University of Technology, 2016, 36(2): 191-196, 214. https://www.cnki.com.cn/Article/CJFDTOTAL-BJLG201602016.htm
[9] 陈志恒, 严利民, 陆斌. 一种快速高效的实时视频去雾算法[J/OL]. 激光与光电子学进展: 1-12. [2020-11-24]. http://kns.cnki.net/kcms/detail/31.1690.TN.20191217.1447.008.html. CHEN Z H, YAN L M, LU B. A fast and efficient real time video defogging algorithm[J/OL]. Laser & Optoelectronics Progress: 1-12. [2020-11-24]. http://kns.cnki.net/kcms/detail/31.1690.TN.20191217.1447.008.html.
计量
- 文章访问数: 294
- HTML全文浏览量: 77
- PDF下载量: 50
