基于多光源模型的夜晚雾天图像去雾算法

doi:10.3969/j.issn.0372-2112.2017.09.011

电子学报 ›› 2017, Vol. 45 ›› Issue (9): 2127-2134.DOI: 10.3969/j.issn.0372-2112.2017.09.011

基于多光源模型的夜晚雾天图像去雾算法

郭璠, 邹北骥, 唐琎

中南大学信息科学与工程学院, 湖南长沙 410083

收稿日期:2016-01-18 修回日期:2016-03-29 出版日期:2017-09-25
- 作者简介:
- 郭璠,女,1982年生于湖南长沙.现为中南大学信息科学与工程学院讲师.主要研究方向为图像处理和模式识别.E-mail:guofancsu@163.com;邹北骥,男,1961年生于江西南昌.现为中南大学信息科学与工程学院教授、博士生导师.主要研究方向为计算机图形学、图像处理、计算机视觉等.E-mail:bjzou@csu.edu.cn;唐琎,男,1966年生于湖南武冈.现为中南大学信息科学与工程学院教授、博士生导师.主要研究方向为图像处理、计算机视觉、模式识别等.E-mail:tjin@csu.edu.cn
- 基金资助:
- 国家自然科学基金 (No.61502537，No.61573380）; 湖南省科技计划重点研发项目 (No.2015WK3006）; 中南大学博士后基金资助项目 (No.126648）

Nighttime Image Defogging Based on Multiple Light Source Model

GUO Fan, ZOU Bei-ji, TANG Jin

School of Information Science and Engineering, Central South University, Changsha, Hunan 410083

Received:2016-01-18 Revised:2016-03-29 Online:2017-09-25 Published:2017-09-25

摘要/Abstract

摘要： 依据夜间多光源导致强光处能见度低的现象，在传统大气散射模型中定义发光因子项，构建了一个专门针对夜晚雾天图像的去雾模型，在此基础上提出了一种夜晚雾天图像去雾算法.该算法将原输入图像分解为新雾天图像层和发光图像层，然后对此分解得到的新雾天图像层进行色偏纠正和引导滤波操作以得到最终的去雾结果.与暗原色原理方法、快速中值滤波方法、图像颜色迁移方法、夜晚成像模型方法等已有方法的对比实验证实了本文算法的有效性.该算法可应用于汽车防碰撞系统、道路监控系统，以及其他识别系统.

关键词: 夜晚, 图像, 去雾, 多光源, 发光图像

Abstract: Multiple light sources cause the low visibility in the brightest region in nighttime images,a glow term was defined in the traditional atmospheric scattering model.A defogging model was constructed for nighttime foggy images.Based on the model,a nighttime image defogging algorithm was proposed.The algorithm decomposed the original image into a nighttime foggy image layer and a glow image layer,and then the color-shift correction operation and the guided filter were applied for the foggy image layer.Thus,the defogging results can be obtained.Comparisons with dark channel prior method,fast median filter method,image color transfer method,and nighttime imaging model method verified the effectiveness of the proposed algorithm.This algorithm can be applied in such devices as vehicle collision avoidance systems,traffic surveillance systems,and other recognition systems.

Key words: nighttime, image, defogging, multiple light source, glow image

中图分类号:

TP399

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.

GUO Fan, ZOU Bei-ji, TANG Jin. Nighttime Image Defogging Based on Multiple Light Source Model[J]. Acta Electronica Sinica, 2017, 45(9): 2127-2134.

参考文献

[1] Shwartz S,Namer E,Schechner Y.Blind haze separation[A].Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].New York,United States:IEEE Computer Society,2006.1984-1991.
[2] Schechner Y Y,Narasimhan S G,Nayar S K.Instant dehazing of images using polarization[A].Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].Los Alamitos,USA:IEEE Computer Society,2001.1325-1332.
[3] Hu W,Yuan G D,Dong Z,Shu X M.Improved single image dehazing using dark channel prior[J].Journal of Computer Research and Development,2010,47(12):2132-2140.
[4] Fattal R.Single image dehazing[J].ACM Transactions on Graphics,2008,27(3):72-79.
[5] Tan R T.Visibility in bad weather from a single image[A].Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C].Anchorage,AK:IEEE Computer Society,2008.1-8.
[6] He K M,Sun J,Tang X O.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.
[7] Tarel J P,Hautiere N.Fast visibility restoration from a single color or gray level image[A].Proceedings of IEEE 12th International Conference on Computer Vision[C].Kyoto,Japan:IEEE Computer Society,2009.2201-2208.
[8] Pei S C,Lee T Y.Nighttime haze removal using color transfer pre-processing and dark channel prior[A].Proceedings of IEEE International Conference on Image Processing[C].Orlando,USA:IEEE Computer Society,2012.957-960.
[9] Zhang J,Cao Y,Wang Z.Nighttime haze removal based on a new imaging model[A].Proceedings of IEEE International Conference on Image Processing[C].Paris,France:IEEE Computer Society,2014.4557-4561.
[10] Lee J W,Lee B H,Won Y,Kim C H,Hong S H.Real-time night visibility enhancement algorithm using the similarity of inverted night image and fog image[J].Lecture Notes in Electrical Engineering,2015,339:1045-1052.
[11] Naik D K,Rout D K.Outdoor image enhancement:increasing visibility under extreme haze and light condition[A].Proceedings of 4th IEEE International Advance Computing Conference[C].Gurgaon,India:IEEE Computer Society,2014.1081-1086.
[12] He K M,Sun J,Tang X O.Guided image filtering[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1397-1409.
[13] Koschmieder H.Theorie der horizontalen Sichtweite:Kontrast und Sichtweite[M].Germany:Keim & Nemnich Press,1925.25-33.
[14] Narasimhan S G,Nayar S K.Shedding light on the weather[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].Madison,USA:IEEE Computer Society,2003.665-672.
[15] Li Y,Brown M S.Single image layer separation using relative smoothness[A].Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C].Columbus,USA:IEEE Computer Society,2014.1-8.
[16] Geman D,Yang C.Nonlinear image recovery with half quadratic regularization[J].IEEE Transactions on Image Processing,1995,4(7):932-946.
[17] Wang Y,Yang J,Yin W,Zhang Y.A new alternating minimization algorithm for total variation image reconstruction[J].SIAM Journal on Imaging Sciences,2008,1(3):248-272.
[18] Hautiere N,Tarel J P,Aubert D,Dumont E.Blind contrast enhancement assessment by gradient ratioing at visible edges[J].Image Analysis and Stereology Journal,2008,27(2):87-95.
[19] Huang K Q,Wang Q,Wu Z Y.Natural color image enhancement and evaluation algorithm based on human visual system[J].Computer Vision and Image Understanding,2006,103(1):52-63.

基于多光源模型的夜晚雾天图像去雾算法

Nighttime Image Defogging Based on Multiple Light Source Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	吴靖, 叶晓晶, 黄峰, 陈丽琼, 王志锋, 刘文犀. 基于深度学习的单帧图像超分辨率重建综述[J]. 电子学报, 2022, 50(9): 2265-2294.
[2]	王波, 王悦, 王伟, 侯嘉尧. 开放环境下自适应聚类优化包络的相机来源取证[J]. 电子学报, 2022, 50(8): 1840-1850.
[3]	张志文, 刘天歌, 聂鹏举. 基于实景数据增强和双路径融合网络的实时街景语义分割算法[J]. 电子学报, 2022, 50(7): 1609-1620.
[4]	徐少平, 李芬, 陈孝国, 陈晓军, 江顺亮. 一种利用改进深度图像先验构建的图像降噪模型[J]. 电子学报, 2022, 50(7): 1573-1578.
[5]	李超, 黄新宇, 王凯. 基于特征融合和自学习锚框的高分辨率图像小目标检测算法[J]. 电子学报, 2022, 50(7): 1684-1695.
[6]	麻文刚, 张亚东, 郭进. 基于景深先验引导与环境光优化的图像去雾[J]. 电子学报, 2022, 50(7): 1708-1721.
[7]	王相海, 赵晓阳, 王鑫莹, 赵克云, 宋传鸣. 非抽取小波边缘学习深度残差网络的单幅图像超分辨率重建[J]. 电子学报, 2022, 50(7): 1753-1765.
[8]	刘杰, 游品鸿, 田明, 刘金凤. 基于局部投影的视差图像拼接平滑优化[J]. 电子学报, 2022, 50(6): 1451-1456.
[9]	曲立国, 陈国豪, 胡俊, 陈鹏. 单次扫描连通域分析算法研究综述[J]. 电子学报, 2022, 50(6): 1521-1536.
[10]	冀振燕, 韩梦豪, 宋晓军, 冯其波. 面向激光光条图像修复的循环相似度映射网络[J]. 电子学报, 2022, 50(5): 1234-1242.
[11]	卓亚琦, 魏家辉, 李志欣. 基于双注意模型的图像描述生成方法研究[J]. 电子学报, 2022, 50(5): 1123-1130.
[12]	佘维, 霍丽娟, 刘炜, 张志鸿, 宋轩, 田钊. 一种可隐藏敏感文档和发送者身份的区块链隐蔽通信模型[J]. 电子学报, 2022, 50(4): 1002-1013.
[13]	杨曦, 张鑫, 郭浩远, 王楠楠, 高新波. 基于不变特征的多源遥感图像舰船目标检测算法[J]. 电子学报, 2022, 50(4): 887-899.
[14]	刘微容, 米彦春, 杨帆, 张彦, 郭宏林, 刘仲民. 基于多级解码网络的图像修复[J]. 电子学报, 2022, 50(3): 625-636.
[15]	王雪松, 张翰林, 程玉虎. 基于自编码器和超图的半监督宽度学习系统[J]. 电子学报, 2022, 50(3): 533-539.