基于CLAHE和图像分解的去雾方法
|
摘要
针对雾天条件下拍摄到的图像对比度低、细节模糊以及颜色暗淡的现象,提出一种基于CLAHE和图像分解的去雾方法。首先,采用限制对比度直方图均衡化(limited contrast histogram equalization,CLAHE)对有雾图像进行增强,有效地提升图像的对比度;然后,在照明—反射模型的基础上,根据照射分量与反射分量的不同特征对增强后的图像进行梯度滤波,将图像进行分解,获得最终包含图像所有细节的反射图像;最后,对反射图像进行Gamma变换,提升图像的亮度,获得最终的去雾图像。利用信息熵、空间频率、平均梯度和运算时间等客观评价标准,与带色彩恢复多尺度Retinex算法(MSRCR算法)和基于暗通道先验去雾算法(He算法)进行对比。实验结果的主观评价和客观评价表明,在雾天图像细节增强和色彩保持方面,本文方法比MSRCR算法和He算法具有更好的效果。
To solve the problems of low contrast,fuzzy and dim color of the images under haze conditions,a new method based on CLAHE and image decomposition is proposed. At first,the foggy image is enhanced by limiting the contrast histogram equalization. Then based on the illuminationreflection model,several gradient filters are used for the enhanced image through the different features between illumination component and reflection component. And the image is decomposed to obtain the reflection image which contains all the details of the image. Finally,a Gamma transform is used for improving the brightness of the reflection component,and get the final fog eliminated image. At the same time,the information entropy,spatial frequency,average gradient and operation time of the objective evaluation criteria are compared with the multi-scale Retinex color restoration algorithm and the dark channel prior to fog eliminating algorithm. The subjective evaluation and objective evaluation of the experimental results show that this method is better than the MSRCR algo-rithm and He's algorithm in the fog image enhancement and color retention.
To solve the problems of low contrast,fuzzy and dim color of the images under haze conditions,a new method based on CLAHE and image decomposition is proposed. At first,the foggy image is enhanced by limiting the contrast histogram equalization. Then based on the illuminationreflection model,several gradient filters are used for the enhanced image through the different features between illumination component and reflection component. And the image is decomposed to obtain the reflection image which contains all the details of the image. Finally,a Gamma transform is used for improving the brightness of the reflection component,and get the final fog eliminated image. At the same time,the information entropy,spatial frequency,average gradient and operation time of the objective evaluation criteria are compared with the multi-scale Retinex color restoration algorithm and the dark channel prior to fog eliminating algorithm. The subjective evaluation and objective evaluation of the experimental results show that this method is better than the MSRCR algo-rithm and He's algorithm in the fog image enhancement and color retention.
引文
[1]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.
[2]张冰冰,戴声奎,孙万源.基于暗原色先验模型的快速去雾算法[J].中国图象图形学报,2013,18(2):184-188.
[3]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):6-11.
[4]蔡超峰,任景英.基于直方图均衡化的手背静脉图像对比度增强[J].计算机应用,2013,33(4):1125-1127.
[5]叶松,赵文昌.基于直方图均衡化的图像增强技术分析与Matlab实现[J].数字技术与应用,2012(1):174-174.
[6]刘德全,崔涛,杨雅宁.局部对比度自适应直方图均衡化图像增强的算法研究[J].信息与电脑,2016(7):79-80.
[7]杨卫中,徐银丽,乔曦,等.基于对比度受限直方图均衡化的水下海参图像增强方法[J].农业工程学报,2016,32(6):197-203.
[8]杨骥,杨亚东,梅雪,等.基于改进的限制对比度自适应直方图的视频快速去雾算法[J].计算机工程与设计,2015,36(1):221-226.
[9]林笑君,梁凤梅.基于Retinex的一种图像去雾算法[J].电视技术,2013,37(17):155-158.
[10]肖胜笔,李燕.具有颜色保真性的快速多尺度Retinex去雾算法[J].计算机工程与应用,2015,51(6):176-180.
[11]JIANG B,WOODELL G A,JOBSON D J.Novel multi-scale retinex with color restoration on graphics processing unit[J].Journal of Real-Time Image Processing,2015,10(2):239-253.
[12]LI S,YEO C,HUA B S.Intrinsic image decomposition using a sparse representation of reflectance[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2013,35(12):2904-2915.
[13]刘轩,刘佳宾.基于对比度受限自适应直方图均衡的乳腺图像增强[J].计算机工程与应用,2008,44(10):173-175.
[14]钟仡龙.单幅图像去雾处理算法研究及软件实现[D].成都:西南交通大学,2012.
[15]阮秋琦.数字图像处理[M].北京:电子工业出版社,2011:72-76.
[16]LI Y,BROWN M S.Single image layer separation using relative smoothness[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.Columbus,USA:IEEE,2014:2752-2759.
[17]LI X,ZHENG S,JIA J.Unnatural L0 sparse representation for natural image deblurring[C]//Proceedings of the IEEE conference on computer vision&pattern recognition.Portland,USA:IEEE,2013:1107-1114.
[18]GEMAN D,YANG C.Nonlinear image recovery with half-quadratic regularization[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,1995,4(7):932-46.
[19]WANG Y,YANG J,YIN W,et al.A New alternating minimization algorithm for total variation image reconstruction[J].Siam Journal on Imaging Sciences,2008,1(3):248-272.
[20]HE K,SUN J,TANG X.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.
[2]张冰冰,戴声奎,孙万源.基于暗原色先验模型的快速去雾算法[J].中国图象图形学报,2013,18(2):184-188.
[3]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):6-11.
[4]蔡超峰,任景英.基于直方图均衡化的手背静脉图像对比度增强[J].计算机应用,2013,33(4):1125-1127.
[5]叶松,赵文昌.基于直方图均衡化的图像增强技术分析与Matlab实现[J].数字技术与应用,2012(1):174-174.
[6]刘德全,崔涛,杨雅宁.局部对比度自适应直方图均衡化图像增强的算法研究[J].信息与电脑,2016(7):79-80.
[7]杨卫中,徐银丽,乔曦,等.基于对比度受限直方图均衡化的水下海参图像增强方法[J].农业工程学报,2016,32(6):197-203.
[8]杨骥,杨亚东,梅雪,等.基于改进的限制对比度自适应直方图的视频快速去雾算法[J].计算机工程与设计,2015,36(1):221-226.
[9]林笑君,梁凤梅.基于Retinex的一种图像去雾算法[J].电视技术,2013,37(17):155-158.
[10]肖胜笔,李燕.具有颜色保真性的快速多尺度Retinex去雾算法[J].计算机工程与应用,2015,51(6):176-180.
[11]JIANG B,WOODELL G A,JOBSON D J.Novel multi-scale retinex with color restoration on graphics processing unit[J].Journal of Real-Time Image Processing,2015,10(2):239-253.
[12]LI S,YEO C,HUA B S.Intrinsic image decomposition using a sparse representation of reflectance[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2013,35(12):2904-2915.
[13]刘轩,刘佳宾.基于对比度受限自适应直方图均衡的乳腺图像增强[J].计算机工程与应用,2008,44(10):173-175.
[14]钟仡龙.单幅图像去雾处理算法研究及软件实现[D].成都:西南交通大学,2012.
[15]阮秋琦.数字图像处理[M].北京:电子工业出版社,2011:72-76.
[16]LI Y,BROWN M S.Single image layer separation using relative smoothness[C]//Proceedings of the IEEE conference on computer vision and pattern recognition.Columbus,USA:IEEE,2014:2752-2759.
[17]LI X,ZHENG S,JIA J.Unnatural L0 sparse representation for natural image deblurring[C]//Proceedings of the IEEE conference on computer vision&pattern recognition.Portland,USA:IEEE,2013:1107-1114.
[18]GEMAN D,YANG C.Nonlinear image recovery with half-quadratic regularization[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,1995,4(7):932-46.
[19]WANG Y,YANG J,YIN W,et al.A New alternating minimization algorithm for total variation image reconstruction[J].Siam Journal on Imaging Sciences,2008,1(3):248-272.
[20]HE K,SUN J,TANG X.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.