当前位置>主页 > 期刊在线 > 计算机技术 >

计算机技术22年17期

基于偏振成像的快速去雾算法
冯晓峰,徐圣奇,刘杰
(中国电子科技集团公司第二十七研究所,河南 郑州 450047)

摘  要:雾霾导致的图像质量退化、图像可视性降低,给交通运输、视频监控乃至军事侦察等带来很大的困难。现有去雾方法的算法复杂度较高,无法满足实时监测或侦察的需求。针对此问题,提出了一种基于偏振成像的快速去雾算法,在大气散射成像模型的基础上,通过合理假设并充分利用偏振探测优势,有效降低了算法复杂度,能够在更短的处理时间内取得与暗通道方法相似的去雾效果,可以广泛应用于实时去雾监测与侦察领域。


关键词:偏振成像;去雾;大气散射成像



DOI:10.19850/j.cnki.2096-4706.2022.17.024


中图分类号:TP301.6                                       文献标识码:A                                     文章编号:2096-4706(2022)17-0094-04


The Fast Defogging Algorithm Based on Polarization Imaging

FENG Xiaofeng, XU Shengqi, LIU Jie

(The 27th Research Institute of China Electronics Technology Group Corporation, Zhengzhou 450047, China)

Abstract: Due to the influence of haze weather, the quality and visibility of image are seriously degraded, which will bring great difficulties to traffic transportation, video surveillance and even military reconnaissance. The existing defogging methods have a higher algorithm complexity and cannot meet the needs of real-time monitoring or reconnaissance. Aiming at this problem, a fast defogging algorithm based on polarization imaging is proposed. On the basis of the atmospheric scattering imaging model, through the reasonable hypothesis and make full use of the advantages of polarization detection, the proposed algorithm can effectively reduce the algorithm complexity and obtain the similar defogging effect with dark channel method in a shorter processing time. It can be widely used in the realtime defogging monitoring and reconnaissance field.

Keywords: polarization imaging; defogging; atmospheric scattering imaging


参考文献:

[1] TAN R T. Visibility in bad weather from a single image [C]//2008 IEEE Conference on Computer Vision and Pattern Recognition.Anchorge,IEEE:2008:1-8.

[2] TAREL J P ,HAUTIÈRE N.Fast visibility restoration from a single color or gray level image [C]//2009 IEEE 12th International Conference on Computer Vision.Kyoto:IEEE,2009:2201-2208.

[3] 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.

[4] 梅康,刘小勤,沐超,等 . 基于自适应指数加权移动平均滤波的快速去雾算法 [J]. 中国激光,2020,47(1):250-259.

[5] 金仙力,张威,刘林峰 . 基于引导滤波和自适应容差的图像去雾算法 [J]. 通信学报,2020,41(5):27-36.

[6] 陈清江,张雪 . 基于并联卷积神经网络的图像去雾 [J]. 自动化学报,2021,47(7):1739-1748.

[7] 冯燕茹 . 双视觉注意网络的联合图像去雾和透射率估计[J]. 光学精密工程,2021,29(4):854-863.

[8] SCHECHNER Y Y,NARASIMHAN S G,NAYAR S K.Instant dehazing of images using polarization [C]//Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.KAUAI:IEEE,2003:I-I.

[9] 代晴晴,范之国,宋强,等 . 全局参数自动估计的彩色图像偏振去雾方法 [J]. 应用光学,2018,39(4):511-517

[10] 夏宏丽,李钢,张仁斌,等 . 基于偏振特性的图像去雾算法 [J]. 计算机应用与软件,2014,31(10):224-226+230.

[11] NAYAR S K,NARASIMHAN S G. Vision in bad weather [C]// Proceedings of the Seventh IEEE International Conference on Computer Vision.Kerkyra:IEEE,2002:820-827.

[12] BASS M. Devices,measurements,and properties [M].McGraw-Hill,1995.

[13] 陈柱铭,郭磊,黄振兴.基于图像增强的去雾算法研究 [J].现代信息科技,2021,5(1):95-98.


作者简介:冯晓峰(1986—),男,汉族,河南平顶山人,工程师,博士,研究方向:数字图像处理、机器学习、人工智能、偏振光探测。