融合暗通道先验和MSRCR的分块调节图像增强算法
|
摘要
针对雾霾天气图像中雾霾浓度分布不均以及色彩失衡等问题,提出一种将暗通道先验算法与带色彩恢复因子的多尺度Retinex算法相结合的分块调节图像增强算法.对大气散射模型以Retinex图像模型中的尺度进行线性映射,得到一个同时具有大气散射模型中的透射率和大气光照值以及Retinex图像模型中入射图像的多参数新模型.根据模型获取去除大气散射图的新原图,并利用不同尺度的引导滤波计算获得整幅图像的入射图像,再结合大气散射光图以及色彩恢复因子得到最终的高频细节图.针对图像中雾霾浓度分布不均的情况将整图划分为多个区域小块,用融合后的算法计算每个区域小块的动态截断值,根据不同的动态截断值可以对整幅图像的高频细节进行动态调整,从而得到多幅局部最优图像,将得到的图像进行像素级等权融合,最后可得到保证各局部细节的最优图像.将本文算法与现有算法在主观视觉和客观评价两方面进行了实验对比,结果表明,该方法可以有效解决图像雾气不均匀以及色彩失衡等问题,明显提高了去雾后图片的质量.
Aiming at the uneven distribution of haze concentration and color imbalance in haze weather images,a block-based image enhancement algorithm is proposed,which combines dark channel prior algorithm with multi-scale retinex with color restore algorithm.The idea of the algorithm is to map the atmospheric scattering model linearly to the scale of the Retinex image model,and get a new multiparameter model with both transmittance and illumination value in the atmospheric scattering model and the incident image in the Retinex image model.The new original image of removing atmospheric scattering image is obtained according to the model,and the incident image of the whole image is calculated by using different scales of guided filtering.Then,the final high-frequency detail image is obtained by combining the atmospheric scattering image and color restoration factor.The whole image is divided into several regions in view of the uneven distribution of haze concentration in the image and the dynamic truncation value of each regional block is calculated by using fusion algorithm.According to different dynamic truncation values,the high frequency details of the whole image can be dynamically adjusted to obtain multiple local optimal images.These images are equal-weighted at the pixel level,and finally the optimal image can be obtained which can retain the local details.The proposed method isexperimentally compared with other existing methods on subjective visual effect and objective evaluation,the results show that the method can effectively solve the problems of non-uniform fog and color imbalance in images,and obviously improve the quality of the image after fog removal.
Aiming at the uneven distribution of haze concentration and color imbalance in haze weather images,a block-based image enhancement algorithm is proposed,which combines dark channel prior algorithm with multi-scale retinex with color restore algorithm.The idea of the algorithm is to map the atmospheric scattering model linearly to the scale of the Retinex image model,and get a new multiparameter model with both transmittance and illumination value in the atmospheric scattering model and the incident image in the Retinex image model.The new original image of removing atmospheric scattering image is obtained according to the model,and the incident image of the whole image is calculated by using different scales of guided filtering.Then,the final high-frequency detail image is obtained by combining the atmospheric scattering image and color restoration factor.The whole image is divided into several regions in view of the uneven distribution of haze concentration in the image and the dynamic truncation value of each regional block is calculated by using fusion algorithm.According to different dynamic truncation values,the high frequency details of the whole image can be dynamically adjusted to obtain multiple local optimal images.These images are equal-weighted at the pixel level,and finally the optimal image can be obtained which can retain the local details.The proposed method isexperimentally compared with other existing methods on subjective visual effect and objective evaluation,the results show that the method can effectively solve the problems of non-uniform fog and color imbalance in images,and obviously improve the quality of the image after fog removal.
引文
[1]LI Yao-yi,DU Yu-chao,GU Zhen-fei.Single image defogging method combined with sky region recognition[J].Computer Engineering and Application,2018,54(19):202-208.李尧羿,杜宇超,顾振飞.结合天空区域识别的单幅图像去雾方法[J].计算机工程与应用,2018,54(19):202-208.
[2]HE Kai-ming,SUN Jian,TANG Xiao-ou.Single image haze removal using dark channel prior[C].IEEE Computer Vision and Pattern Recognition,2009:1956-1963.
[3]MENG Gao-feng,WANG Ying,DUAN Jiang-yong,et al.Efficient image dehazing with boundary constraint and contextual regularization[C].2013IEEE International Conference on Computer Vision,2013:617-624.
[4]MATAN S,GLATZER I,FATTAL R,et al.Automatic recovery of the atmospheric light in hazy images[C].2014IEEEInternational Conference on Computational Photography,2014:1-11.
[5]SINGH G,JAGGI N,VASAMSETTI S,et al.Underwater image/video enhancement using wavelet based color correction(WBCC)method[C].2015IEEE Underwater Technology,2015:1-5.
[6]STARCK J L,MURTAGH F,CANDES E J,et al.Gray and color image contrast enhancement by the curvelet transform[J].IEEE Transactions on Image Processing,2003,12(6):706-717.
[7]LIU Hao-ting,LU Han-qing,ZHANG Yu.Image enhancement for outdoor long-range surveillance using IQ-learning multiscale Retinex[J].IET Image Processing,2017,11(9):786-795.
[8]FU Q,JUNG C,XU K,et al.Retinex-based perceptual contrast enhancement in images using luminance adaptation[J].IEEE Access,2018,(6):61277-61286.
[9]WU He-nan,YANG Guang,XU Zhong-lin,et al.Remote sensing image enhancement method based on multi-scale retinex[C].IEEE International Conference of Information Technology,Computer Engineering and Management Sciences,2011(3):15-18.
[10]CUI Ying.Image defogging method based on histogram[D].Changchun:Jilin University,2018.崔莹.基于直方图的图像去雾方法[D].长春:吉林大学,2018.
[11]WANG Yong-xin,DIAO Ming,HAN Chuang.Underwater imaging enhancement algorithm under conventional light source based on iterative histogram equalization[J].Acta Photonica Sinica,2018,47(11):1101002.王永鑫,刁鸣,韩闯.基于迭代直方图均衡化的常规光源下水下成像增强算法[J].光子学报,2018,47(11):1101002.
[12]SINGH K,KAPOOR R.Image enhancement via median-mean based sub-image-clipped histogram equalization[J].OptikInternational Journal for Light and Electron Optics,2014,125(17):4646-4651.
[13]CUI Xu-dong,YANG You.Traffic haze image enhancement based on HE and improved MSRCR[J].Journal of Chongqing Normal University(Natural Science Edition),2018,35(1):100-106.崔旭东,杨有.结合HE和改进MSRCR的交通雾霾图像增强[J].重庆师范大学学报(自然科学版),2018,35(1):100-106.
[14]ZHAO Chun-li,DONG Jing-wei.Enhancement algorithm of haze image based on dark channel and multi-scale Retinex[J].Laser Magazine,2018,39(1):104-109.赵春丽,董静薇.基于暗通道及多尺度Retinex的雾霾天气图像增强算法[J].激光杂志,2018,39(1):104-109.
[15]ZHOU Jie.Dark channel prior image and Retinex theory to research based on fog[D].Lanzhou:Lanzhou Jiaotong University,2018.周杰.基于暗通道先验和Retinex理论的图像去雾研究[D].兰州:兰州交通大学,2018.
[16]SHI Lei,LI Yang,CHU Shi-bo,et al.Image haze removal using dark channel prior and minimizing energy function[C].IEEE 2nd Information Technology,Networking,Electronic and Automation Control Conference,2017:256-259.
[17]JI Xiao-qiang.Research on image fast defogging and sharpness restoration technology[D].Beijing:Graduate School of Chinese Academy of Sciences,2012.嵇晓强.图像快速去雾与清晰度恢复技术研究[D].北京:中国科学院研究生院,2012.
[18]LI Yi.Research on infrared image enhancement technology based on visual model[D].Beijing:Graduate School of Chinese Academy of Sciences,2016.李毅.基于视觉模型的红外图像增强技术研究[D].北京:中国科学院研究生院,2016.
[19]LIN Jia-biao.Research on restoration and enhancement algorithm of single fog image[D].Ganzhou:Jiangxi University of Technology,2013.林家彪.单幅雾天图像的复原和增强算法研究[D].赣州:江西理工大学,2013.
[20]HE Kai-ming,SUN Jian,TANG Xiao-ou.Guided image filtering[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1397-1409.
[21]FANG Shuai,YANG Jing-rong,CAO Yang,et al.Local multi-scale Retinex algorithm for image-guided filtering[J].Chinese Journal of Image Graphics,2012,17(7):748-755.方帅,杨静荣,曹洋,等.图像引导滤波的局部多尺度Retinex算法[J].中国图象图形学报,2012,17(7):748-755.
[22]ZHAN Kun,SHI Jin-hui,TENG Ji-cai,et al.Linking synaptic computation for image enhancement[J].Neurocomputing,2017,(238):1-12.
[23]CHO Y G,JEONG J Y,KIM A,et al.Model-assisted multiband fusion for single image enhancement and applications to robot vision[J].IEEE Robotics and Automation Letters,2018,3(4):2822-2829.
[24]MINNEN D,TODERICI G,SINGH S,et al.Image-dependent local entropy models for learned image compression[C].IEEE International Conference on Image Processing,2018:430-434.
[25]KISH L B,NIKLASSON G A,GRANQVIST C G,et al.Facts and myths about zero-point thermal noise,and information entropy versus thermal entropy[C].2017International Conference on Noise and Fluctuations,2017:1-2.
[26]HUANG Wen-hui,CHEN Ren-lei,ZHANG Jia-mou.Improvement and realization of objective measurement method for digital video image quality[J].Journal of Beijing University of Posts and Telecommunications,2005,28(4):87-90.黄文辉,陈仁雷,张家谋.数字视频图像质量客观测量方法的改进与实现[J].北京邮电大学学报,2005,28(4):87-90.
[27]HAUTIERE N,TAREL J P,DIDIER A,et al.Blind contrast enhancement assessment by gradient ratioing at visible edges[J].Image Analysis and Stereology,2008,27(2):87-95.
[2]HE Kai-ming,SUN Jian,TANG Xiao-ou.Single image haze removal using dark channel prior[C].IEEE Computer Vision and Pattern Recognition,2009:1956-1963.
[3]MENG Gao-feng,WANG Ying,DUAN Jiang-yong,et al.Efficient image dehazing with boundary constraint and contextual regularization[C].2013IEEE International Conference on Computer Vision,2013:617-624.
[4]MATAN S,GLATZER I,FATTAL R,et al.Automatic recovery of the atmospheric light in hazy images[C].2014IEEEInternational Conference on Computational Photography,2014:1-11.
[5]SINGH G,JAGGI N,VASAMSETTI S,et al.Underwater image/video enhancement using wavelet based color correction(WBCC)method[C].2015IEEE Underwater Technology,2015:1-5.
[6]STARCK J L,MURTAGH F,CANDES E J,et al.Gray and color image contrast enhancement by the curvelet transform[J].IEEE Transactions on Image Processing,2003,12(6):706-717.
[7]LIU Hao-ting,LU Han-qing,ZHANG Yu.Image enhancement for outdoor long-range surveillance using IQ-learning multiscale Retinex[J].IET Image Processing,2017,11(9):786-795.
[8]FU Q,JUNG C,XU K,et al.Retinex-based perceptual contrast enhancement in images using luminance adaptation[J].IEEE Access,2018,(6):61277-61286.
[9]WU He-nan,YANG Guang,XU Zhong-lin,et al.Remote sensing image enhancement method based on multi-scale retinex[C].IEEE International Conference of Information Technology,Computer Engineering and Management Sciences,2011(3):15-18.
[10]CUI Ying.Image defogging method based on histogram[D].Changchun:Jilin University,2018.崔莹.基于直方图的图像去雾方法[D].长春:吉林大学,2018.
[11]WANG Yong-xin,DIAO Ming,HAN Chuang.Underwater imaging enhancement algorithm under conventional light source based on iterative histogram equalization[J].Acta Photonica Sinica,2018,47(11):1101002.王永鑫,刁鸣,韩闯.基于迭代直方图均衡化的常规光源下水下成像增强算法[J].光子学报,2018,47(11):1101002.
[12]SINGH K,KAPOOR R.Image enhancement via median-mean based sub-image-clipped histogram equalization[J].OptikInternational Journal for Light and Electron Optics,2014,125(17):4646-4651.
[13]CUI Xu-dong,YANG You.Traffic haze image enhancement based on HE and improved MSRCR[J].Journal of Chongqing Normal University(Natural Science Edition),2018,35(1):100-106.崔旭东,杨有.结合HE和改进MSRCR的交通雾霾图像增强[J].重庆师范大学学报(自然科学版),2018,35(1):100-106.
[14]ZHAO Chun-li,DONG Jing-wei.Enhancement algorithm of haze image based on dark channel and multi-scale Retinex[J].Laser Magazine,2018,39(1):104-109.赵春丽,董静薇.基于暗通道及多尺度Retinex的雾霾天气图像增强算法[J].激光杂志,2018,39(1):104-109.
[15]ZHOU Jie.Dark channel prior image and Retinex theory to research based on fog[D].Lanzhou:Lanzhou Jiaotong University,2018.周杰.基于暗通道先验和Retinex理论的图像去雾研究[D].兰州:兰州交通大学,2018.
[16]SHI Lei,LI Yang,CHU Shi-bo,et al.Image haze removal using dark channel prior and minimizing energy function[C].IEEE 2nd Information Technology,Networking,Electronic and Automation Control Conference,2017:256-259.
[17]JI Xiao-qiang.Research on image fast defogging and sharpness restoration technology[D].Beijing:Graduate School of Chinese Academy of Sciences,2012.嵇晓强.图像快速去雾与清晰度恢复技术研究[D].北京:中国科学院研究生院,2012.
[18]LI Yi.Research on infrared image enhancement technology based on visual model[D].Beijing:Graduate School of Chinese Academy of Sciences,2016.李毅.基于视觉模型的红外图像增强技术研究[D].北京:中国科学院研究生院,2016.
[19]LIN Jia-biao.Research on restoration and enhancement algorithm of single fog image[D].Ganzhou:Jiangxi University of Technology,2013.林家彪.单幅雾天图像的复原和增强算法研究[D].赣州:江西理工大学,2013.
[20]HE Kai-ming,SUN Jian,TANG Xiao-ou.Guided image filtering[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1397-1409.
[21]FANG Shuai,YANG Jing-rong,CAO Yang,et al.Local multi-scale Retinex algorithm for image-guided filtering[J].Chinese Journal of Image Graphics,2012,17(7):748-755.方帅,杨静荣,曹洋,等.图像引导滤波的局部多尺度Retinex算法[J].中国图象图形学报,2012,17(7):748-755.
[22]ZHAN Kun,SHI Jin-hui,TENG Ji-cai,et al.Linking synaptic computation for image enhancement[J].Neurocomputing,2017,(238):1-12.
[23]CHO Y G,JEONG J Y,KIM A,et al.Model-assisted multiband fusion for single image enhancement and applications to robot vision[J].IEEE Robotics and Automation Letters,2018,3(4):2822-2829.
[24]MINNEN D,TODERICI G,SINGH S,et al.Image-dependent local entropy models for learned image compression[C].IEEE International Conference on Image Processing,2018:430-434.
[25]KISH L B,NIKLASSON G A,GRANQVIST C G,et al.Facts and myths about zero-point thermal noise,and information entropy versus thermal entropy[C].2017International Conference on Noise and Fluctuations,2017:1-2.
[26]HUANG Wen-hui,CHEN Ren-lei,ZHANG Jia-mou.Improvement and realization of objective measurement method for digital video image quality[J].Journal of Beijing University of Posts and Telecommunications,2005,28(4):87-90.黄文辉,陈仁雷,张家谋.数字视频图像质量客观测量方法的改进与实现[J].北京邮电大学学报,2005,28(4):87-90.
[27]HAUTIERE N,TAREL J P,DIDIER A,et al.Blind contrast enhancement assessment by gradient ratioing at visible edges[J].Image Analysis and Stereology,2008,27(2):87-95.