基于明亮区域和天空识别的图像去雾算法
|
摘要
针对去雾算法对存在大面积明亮区域的图像去雾效果不佳的问题,根据暗原色先验原理提出了一种基于明亮区域和天空区域识别的图像去雾算法。根据天空区域和明亮区域的特点,建立了明亮区域和天空区域的判别机制,对不同类型的图像采用不同方法求取大气光值;基于点暗原色原理,采用中值滤波和双边滤波对点透射率进行细化,与块透射率相比,采用点透射率可以保留更多的图像细节。结果显示,采用均方误差、峰值信噪比、可见边梯度和结构相似性4个参数作为图像去雾效果的客观评价指标,优化后算法的综合评价指标均优于其他算法。所提出的算法对不同类型图像均可取得较好的去雾效果,消除了去雾图像的光晕、块效应和天空区域过饱和现象。
The existing image dehazing algorithms cannot achieve satisfactory performance in the cases that the image has large high light regions. To solve this problem, an improved image haze removal method is proposed based on sky region and high light region detection. Discriminant mechanism for sky region and high light region is established, and different methods are used to calculate the atmospheric light value for different types of images. Based on the principle of point dark channel, the transmission is further detailed by median filtering and bilateral filtering in order to retain more image details. Finally, the mean square error(MSE), peak signal to noise ratio(PSNR), visible edge gradient method and structure similarit(SSIM) are used as the objective criterion to judge the dehazing effect. The calculation results show that the optimized algorithm is superior to other algorithms in the aspect of comprehensive evaluation index. The proposed method has better performance in haze removal for different types of images, effectively eliminating the image halo, block effect and over saturation of the sky.
The existing image dehazing algorithms cannot achieve satisfactory performance in the cases that the image has large high light regions. To solve this problem, an improved image haze removal method is proposed based on sky region and high light region detection. Discriminant mechanism for sky region and high light region is established, and different methods are used to calculate the atmospheric light value for different types of images. Based on the principle of point dark channel, the transmission is further detailed by median filtering and bilateral filtering in order to retain more image details. Finally, the mean square error(MSE), peak signal to noise ratio(PSNR), visible edge gradient method and structure similarit(SSIM) are used as the objective criterion to judge the dehazing effect. The calculation results show that the optimized algorithm is superior to other algorithms in the aspect of comprehensive evaluation index. The proposed method has better performance in haze removal for different types of images, effectively eliminating the image halo, block effect and over saturation of the sky.
引文
[1] 吴迪,朱青松.图像去雾的最新研究进展[J].自动化学报,2015,41(2):221-239.WU Di,ZHU Qingsong.The latest research progress of image dehazing[J].Acta Automatica Sinica,2015,41(2):221-239.
[2] TAREL J P,HAUTIERE N.Fast visibility restoration from a single color or gray level image[C]//IEEE 12th International Conference on Computer Vision.Kyoto:[s.n.],2009:2201-2208.
[3] GAO Yuanyuan,HU Haimiao,WANG Shuhang,et al.A fast image dehazing algorithm based on negative correction[J].Signal Processing,2014,103 (10):380-398.
[4] NARASIMHAN S G,NAYAR S K.Contrast restoration of weather degraded images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(6):713-724.
[5] SHWARTZ S,NAMER E,SCHECHNER Y Y.Blind haze separation[C]//Proceedings of IEEE Computer Vision and Pattern Recognition.[S.l.]:[s.n.],2006:1984-1991.
[6] TAN R T.Visibility in bad weather from a single image[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Anchorage:[s.n.],2008:1-8.
[7] FATTAL R.Single image dehazing[J].ACM Trans on Grapics,2008,27(3):1-9.
[8] HE Kaiming,SUN Jian,TANG Xiaoou.Single image haze removal using dark channel prior[C]//IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:[s.n.],2011:1956-1963.
[9] 蒋建国,侯天峰,齐美彬.改进的基于暗原色先验的图像去雾算法[J].电路与系统学报,2011,16(2):7-12.JIANG Jianguo,HOU Tianfeng,QI Meibin.Improved algorithm on image haze remova1 using dark channel prior[J].Journal of Circuits and Systems,2011,16(2):7-12.
[10] 曾接贤,余永龙.双边滤波与暗通道结合的图像保边去雾算法[J].中国图象图形学报,2017,22 (2):147-153.ZENG Jiexian,YU Yonglong.Image defogging and edge preserving algorithm based on dark channel prior and bilateral filtering[J].Journal of Image and Graphics,2017,22(2):147-153.
[11] KIM J H,JANG W D,SIM J Y,et al.Optimized contrast enhancement for real-time image and video dehazing[J].Journal of Visual Communication and Image Representation,2013,24(3):410- 425.
[12] MCCARTNEY E J,HALL F F.Optics of the atmosphere:Scattering by molecules and particles[J].Physics Today,1971,30(5):76.
[13] 王殿伟,闫伟超,刘颖,等,基于暗原色先验的快速单幅图像去雾算法[J],计算机应用研究,2018,35(12):3836- 3840.WANG Dianwei,YAN Weichao,LIU Ying,et al.Fast single image dehazing algorithm based on dark channel prior[J].Application Research of Computers,2018,35(12):3836-3840.
[14] 禹晶,李大鹏,廖庆敏.基于物理模型的快速单幅图像去雾方法[J].自动化学报,2011,37(2):143-149.YU Jing,LI Dapeng,LIAO Qingmin.Physics-based fast single image fog removal[J].Acta Automatica Sinica,2011,37(2):143-149.
[15] WANG Zhongliang,FENG Yan.Fast single haze image enhancement[J].Computers and Electrical Engineering,2014,40(3):785-795.
[16] 王一帆,尹传历,黄义明,等,基于双边滤波的图像去雾[J].中国图象图形学报,2014,19(3):386-392.WANG Yifan,YIN Chuanli,HUANG Yiming,et al.Image haze removal using a bilateral filter[J].Journal of Image and Graphics,2014,19(3):386-392.
[17] BISWAS B,ROY P,CHOUDHURI R,et al.Microscopic image contrast and brightness enhancement using multi-scale retinex and cuckoo search algorithm [J].Procedia Computer Science,2015,70:348-354.
[2] TAREL J P,HAUTIERE N.Fast visibility restoration from a single color or gray level image[C]//IEEE 12th International Conference on Computer Vision.Kyoto:[s.n.],2009:2201-2208.
[3] GAO Yuanyuan,HU Haimiao,WANG Shuhang,et al.A fast image dehazing algorithm based on negative correction[J].Signal Processing,2014,103 (10):380-398.
[4] NARASIMHAN S G,NAYAR S K.Contrast restoration of weather degraded images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(6):713-724.
[5] SHWARTZ S,NAMER E,SCHECHNER Y Y.Blind haze separation[C]//Proceedings of IEEE Computer Vision and Pattern Recognition.[S.l.]:[s.n.],2006:1984-1991.
[6] TAN R T.Visibility in bad weather from a single image[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Anchorage:[s.n.],2008:1-8.
[7] FATTAL R.Single image dehazing[J].ACM Trans on Grapics,2008,27(3):1-9.
[8] HE Kaiming,SUN Jian,TANG Xiaoou.Single image haze removal using dark channel prior[C]//IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:[s.n.],2011:1956-1963.
[9] 蒋建国,侯天峰,齐美彬.改进的基于暗原色先验的图像去雾算法[J].电路与系统学报,2011,16(2):7-12.JIANG Jianguo,HOU Tianfeng,QI Meibin.Improved algorithm on image haze remova1 using dark channel prior[J].Journal of Circuits and Systems,2011,16(2):7-12.
[10] 曾接贤,余永龙.双边滤波与暗通道结合的图像保边去雾算法[J].中国图象图形学报,2017,22 (2):147-153.ZENG Jiexian,YU Yonglong.Image defogging and edge preserving algorithm based on dark channel prior and bilateral filtering[J].Journal of Image and Graphics,2017,22(2):147-153.
[11] KIM J H,JANG W D,SIM J Y,et al.Optimized contrast enhancement for real-time image and video dehazing[J].Journal of Visual Communication and Image Representation,2013,24(3):410- 425.
[12] MCCARTNEY E J,HALL F F.Optics of the atmosphere:Scattering by molecules and particles[J].Physics Today,1971,30(5):76.
[13] 王殿伟,闫伟超,刘颖,等,基于暗原色先验的快速单幅图像去雾算法[J],计算机应用研究,2018,35(12):3836- 3840.WANG Dianwei,YAN Weichao,LIU Ying,et al.Fast single image dehazing algorithm based on dark channel prior[J].Application Research of Computers,2018,35(12):3836-3840.
[14] 禹晶,李大鹏,廖庆敏.基于物理模型的快速单幅图像去雾方法[J].自动化学报,2011,37(2):143-149.YU Jing,LI Dapeng,LIAO Qingmin.Physics-based fast single image fog removal[J].Acta Automatica Sinica,2011,37(2):143-149.
[15] WANG Zhongliang,FENG Yan.Fast single haze image enhancement[J].Computers and Electrical Engineering,2014,40(3):785-795.
[16] 王一帆,尹传历,黄义明,等,基于双边滤波的图像去雾[J].中国图象图形学报,2014,19(3):386-392.WANG Yifan,YIN Chuanli,HUANG Yiming,et al.Image haze removal using a bilateral filter[J].Journal of Image and Graphics,2014,19(3):386-392.
[17] BISWAS B,ROY P,CHOUDHURI R,et al.Microscopic image contrast and brightness enhancement using multi-scale retinex and cuckoo search algorithm [J].Procedia Computer Science,2015,70:348-354.