基于改进多尺度Retinex的单幅彩色图像增强算法
|
摘要
针对传统基于多尺度Retinex的单幅彩色图像增强算法,进行图像增强时未考虑亮度分量和局部对比度,导致图像增强未达到最佳尺度值,获取的增强图像存在失真、带有雾膜的问题,提出一种基于改进多尺度Retinex的单幅彩色图像增强算法.首先采用高斯环境函数对3种灰度图像实施卷积并进行灰度纠正,将3种纠正灰度后的图像合并,获取初步单幅彩色图像增强结果;然后改进多尺度Retinex的单幅彩色图像增强算法,增强单幅彩色图像像素的平均亮度;最后通过非线性函数对亮度增强后的彩色图像实施局部对比度增强处理,用线性调整方法恢复局部对比度增强后的图像颜色,以获取最佳单幅彩色图像的图像增强结果.实验结果表明,该算法可获得亮度和对比度均较好的单幅彩色图像增强结果.
Aiming at the problem that the traditional single color image enhancement algorithm based on multi-scale Retinex did not consider the brightness component and local contrast when image enhancement was performed,which led to the image enhancement not reaching the optimal scale value,and the obtained enhanced image had the distortion and fog film,we proposed a single color image enhancement algorithm based on the improved multi-scale Retinex.Firstly,the Gaussian environment function was used to convolve the three grayscale images and perform grayscale correction.The three corrected grayscale images were combined to obtain a preliminary single color image enhancement result.Secondly,we improved multi-scale Retinex single color image enhancement algorithm to enhance the average brightness of single color image pixels.Finally,local contrast enhancement processing of color image after brightness enhanced was performed by nonlinear function,and image color after local contrast enhancement was restored by linear adjustment method to obtain the best image enhancement results of single color image.The experimental results show that the proposed algorithm can obtain single color image enhancement results with good brightness and contrast.
Aiming at the problem that the traditional single color image enhancement algorithm based on multi-scale Retinex did not consider the brightness component and local contrast when image enhancement was performed,which led to the image enhancement not reaching the optimal scale value,and the obtained enhanced image had the distortion and fog film,we proposed a single color image enhancement algorithm based on the improved multi-scale Retinex.Firstly,the Gaussian environment function was used to convolve the three grayscale images and perform grayscale correction.The three corrected grayscale images were combined to obtain a preliminary single color image enhancement result.Secondly,we improved multi-scale Retinex single color image enhancement algorithm to enhance the average brightness of single color image pixels.Finally,local contrast enhancement processing of color image after brightness enhanced was performed by nonlinear function,and image color after local contrast enhancement was restored by linear adjustment method to obtain the best image enhancement results of single color image.The experimental results show that the proposed algorithm can obtain single color image enhancement results with good brightness and contrast.
引文
[1]高银,云利军,石俊生,等.基于各向异性高斯滤波的暗原色理论雾天彩色图像增强算法[J].计算机辅助设计与图形学学报,2015,27(9):1701-1706.(GAO Yin,YUN Linjun,SHI Junsheng,et al.Enhancement Dark Channel Algorithm of Color Fog Image Based on the Anisotropic Gaussian Filtering[J].Journal of ComputerAided Design and Graphics,2015,27(9):1701-1706.)
[2]程新,刘一博,张红.图像通信中一种低照度彩色图像增强算法[J].电子器件,2016,39(6):1338-1342.(CHENG Xin,LIU Yibo,ZHANG Hong.A Low Illumination Color Image Enhancement Algorithm in Image Communication[J].Chinese Journal of Electronic Device,2016,39(6):1338-1342.)
[3]曹风云,赵凯,王筱薇倩,等.自适应水下彩色图像增强算法[J].电子测量与仪器学报,2016,30(5):772-778.(CAO Fengyun,ZHAO Kai,WANG Xiaoweiqian,et al.An Adaptive Underwater Image Enhancement Algorithm[J].Journal of Electronic Measurement and Instrument,2016,30(5):772-778.)
[4]李益红,周晓谊.一种多分辨多尺度的Retinex彩色图像增强算法[J].计算机工程与应用,2017,53(16):193-198.(LI Yihong,ZHOU Xiaoyi.Multi-resolution and Multi-scale Retinex for Color Image Enhancement[J].Computer Engineering and Applications,2017,53(16):193-198.)
[5]LI Hui,XIE Wenhao,WANG Xingang,et al.GPU Implementation of Multi-scale Retinex Image Enhancement Algorithm[C]//2016 IEEE/ACS 13th International Conference of Computer Systems and Applications.Piscataway,NJ:IEEE,2017:1-5.
[6]温海滨,毕笃彦,马时平,等.消除光晕和细节增强的多尺度Retinex红外图像增强[J].红外技术,2016,38(2):149-156.(WEN Haibin,BI Duyan,MA Shiping,et al.Halo-Free and Detail Enhancement Based on Multiscale Retinex for Infrared Image[J].Infrared Technology,2016,38(2):149-156.)
[7]MEI Xue,YANG Ji,ZHANG Yuyan,et al.Video Image Dehazing Algorithm Based on Multi-scale Retinex with Color Restoration[C]//International Conference on Smart Grid and Electrical Automation.Piscataway,NJ:IEEE,2016:195-200.
[8]唐宁,赵鹏,吴绍启.改进多尺度Retinex的彩色图像增强[J].电子设计工程,2016,24(12):168-171.(TANGNing,ZHAO Peng,WU Shaoqi.Color Image Enhancement Based on Improved Multi-scale Retinex[J].Electronic Design Engineering,2016,24(12):168-171.)
[9]YU Tianhe,MENG Xue,ZHU Ming,et al.An Improved Multi-scale Retinex Fog and Haze Image Enhancement Method[C]//International Conference on Information System and Artificial Intelligence.Piscataway,NJ:IEEE,2017:557-560.
[2]程新,刘一博,张红.图像通信中一种低照度彩色图像增强算法[J].电子器件,2016,39(6):1338-1342.(CHENG Xin,LIU Yibo,ZHANG Hong.A Low Illumination Color Image Enhancement Algorithm in Image Communication[J].Chinese Journal of Electronic Device,2016,39(6):1338-1342.)
[3]曹风云,赵凯,王筱薇倩,等.自适应水下彩色图像增强算法[J].电子测量与仪器学报,2016,30(5):772-778.(CAO Fengyun,ZHAO Kai,WANG Xiaoweiqian,et al.An Adaptive Underwater Image Enhancement Algorithm[J].Journal of Electronic Measurement and Instrument,2016,30(5):772-778.)
[4]李益红,周晓谊.一种多分辨多尺度的Retinex彩色图像增强算法[J].计算机工程与应用,2017,53(16):193-198.(LI Yihong,ZHOU Xiaoyi.Multi-resolution and Multi-scale Retinex for Color Image Enhancement[J].Computer Engineering and Applications,2017,53(16):193-198.)
[5]LI Hui,XIE Wenhao,WANG Xingang,et al.GPU Implementation of Multi-scale Retinex Image Enhancement Algorithm[C]//2016 IEEE/ACS 13th International Conference of Computer Systems and Applications.Piscataway,NJ:IEEE,2017:1-5.
[6]温海滨,毕笃彦,马时平,等.消除光晕和细节增强的多尺度Retinex红外图像增强[J].红外技术,2016,38(2):149-156.(WEN Haibin,BI Duyan,MA Shiping,et al.Halo-Free and Detail Enhancement Based on Multiscale Retinex for Infrared Image[J].Infrared Technology,2016,38(2):149-156.)
[7]MEI Xue,YANG Ji,ZHANG Yuyan,et al.Video Image Dehazing Algorithm Based on Multi-scale Retinex with Color Restoration[C]//International Conference on Smart Grid and Electrical Automation.Piscataway,NJ:IEEE,2016:195-200.
[8]唐宁,赵鹏,吴绍启.改进多尺度Retinex的彩色图像增强[J].电子设计工程,2016,24(12):168-171.(TANGNing,ZHAO Peng,WU Shaoqi.Color Image Enhancement Based on Improved Multi-scale Retinex[J].Electronic Design Engineering,2016,24(12):168-171.)
[9]YU Tianhe,MENG Xue,ZHU Ming,et al.An Improved Multi-scale Retinex Fog and Haze Image Enhancement Method[C]//International Conference on Information System and Artificial Intelligence.Piscataway,NJ:IEEE,2017:557-560.