小波和傅里叶相融合的彩色木材图像增强
|
摘要
针对现实生活中彩色图像普遍存在不清晰和对比度差的情况,在RGB模型上提出了一种新的彩色图像增强算法,并且应用到了木材图像领域。将彩色木材图像分解为RGB 3个通道,首先使用滤波器把3个通道分别分解成高低频子带;然后使用傅立叶变换和小波变换相融合的方式进行锯齿检测,同时进行阈值判断;之后对检测到的锯齿进行消除,低频子带使用方向自适应滤波器,高频子带使用小波收缩函数进行消除;再使用小波逆变换返回3个通道;最后将3个通道还原成彩色图像。结果表明,该方法和传统方法相比较,可以有效保持图像的边缘特征,达到增强效果。以樟子松微观横截面为例,峰值信噪比PSNR提高了5.05,信息熵提高了3.14。本研究同时采集了榆木微观横截面、杨木宏观横截面和云杉微观横截面,其图像均得到增强。
With the popularization of color image,people generally suffer from its unclearness and poor contrast in real life.In this paper,a new color image enhancement algorithm was proposed based on the RGB model,and it was then applied to the wood images.The color wood image was decomposed into three RGB channels.First,a filter was used to decompose the three channels into high-and low-frequency subbands; then the jagged detection was performed by using the Fourier transform and the wavelet transform,and the threshold value was judged.The saw tooth detected were eliminated,the low-and high frequency subbands were eliminated by a direction-adaptive filter,and a wavelet contraction function,respectively; then the wavelet inverse transform was used to return three channels; finally the three channels were restored to a color image.Compared with the traditional method,this method could effectively maintain the edge features of the image and achieve the enhanced effect.Taking the Pinus sylvestris microscopic cross section as an example,the peak signal-to-noise ratio(PSNR) increased by 5.05 and the information entropy increased by 3.14.The microscopic cross sections of eucalyptus,the macro cross section of poplar,and the microscopic cross section of spruce were also collected to increase the experimental results.
With the popularization of color image,people generally suffer from its unclearness and poor contrast in real life.In this paper,a new color image enhancement algorithm was proposed based on the RGB model,and it was then applied to the wood images.The color wood image was decomposed into three RGB channels.First,a filter was used to decompose the three channels into high-and low-frequency subbands; then the jagged detection was performed by using the Fourier transform and the wavelet transform,and the threshold value was judged.The saw tooth detected were eliminated,the low-and high frequency subbands were eliminated by a direction-adaptive filter,and a wavelet contraction function,respectively; then the wavelet inverse transform was used to return three channels; finally the three channels were restored to a color image.Compared with the traditional method,this method could effectively maintain the edge features of the image and achieve the enhanced effect.Taking the Pinus sylvestris microscopic cross section as an example,the peak signal-to-noise ratio(PSNR) increased by 5.05 and the information entropy increased by 3.14.The microscopic cross sections of eucalyptus,the macro cross section of poplar,and the microscopic cross section of spruce were also collected to increase the experimental results.
引文
[1] 冯清枝,虎添翼.一种基于累积分布的低照度彩色图像增强算法[J].光电技术应用,2017,32(5):57-61.FENG Q Z,HU T Y.A novel algorithm based on cumulative distribution for low illumination color image enhancement[J].Electro-optic Technology Application,2017,32(5):57-61.(in Chinese)
[2] 安静,张贵仓,刘燕妮.基于多尺度top-hat变换的自适应彩色图像增强[J].计算机工程与科学,2017,3(7):1317-1321.AN J,ZHANG G C,LIU Y N.Adaptive color image enhancement based on multi-scale top-hat transform[J].Computer Engineering&Science,2017,3(7):1317-1321.(in Chinese)
[3] 杨艳,陈章宝,王艳春.基于四元数傅里叶变换的彩色图像增强[J].蚌埠学院学报,2017,6(6):9-12.YANG Y,CHEN Z B,WANG Y C.Color images enhancing based on quaternion fourier transforms[J].Journal of Benghu University,2017,6(6):9-12.(in Chinese)
[4] 何惜琴,许艳华.一种改进的仿生彩色图像增强算法[J].闽江学院学报,2017(5):49-55.HENG X Q,XU Y H.An improved bio-inspired algorithm for color image enhancement[J].Journal of Minjiang University,2017(5):49-55.(in Chinese)
[5] JOBSON D J,RAHMA N Z,WOODLL G A.Properties and perform-ance of a center/surround retinex[J].IEEE Transactions on Image Processing,1997,6(3):451-462.
[6] RAHMAN Z U,JOBSON D J,WOODELL G A.Retinex processing for automatic images enhancement[J].Journal of Electrontic Imaging,2004,13(1):100-110.
[7] 蔡式东,杨芳.基于直方图修正的图像增强算法[J].光电子技术,2016.9,32(3):155-160.CAI S D,YANG F.Image enhancement based on histogram modification[J].Opto Electronic Technology,2016,9,32(3):155-160.(in Chinese)
[8] 韩磊,曲中水.一种RGB模型彩色图像增强方法[J].哈尔滨理工大学学报,2014,19(6):59-64.HAN L,QU Z S.A kind of enhancement method of color image based on RGB color model[J].Journal of Harbin University of Science And Technology,2014,19(6):59-64.(in Chinese)
[9] GAN J.Non-separable wavelet-like lifting structure for image and video compression with aliasing suppression[J].IEEE International Conference on Image Processing,San Antonio,USA,2007.10(6):65-68.
[10] 杜颖财,宋路,万秋华.小波变换实现的光电编码器精确实时测速[J].红外与激光工程,2017,46(5):1-6.DU Y C,SONG L,WAN Q H.Accurate real-time velocity measurement of photoelectric encoder based on wavelet transform[J].Infrared and Laser Engineering,2017,46(5):1-6.(in Chinese)
[11] COULANGE B.An aliasing detection algorithm based on suspicious colocalizations of fourier coefficients[J].IEEE International Conference on Image Processing,HongKong,China,2010(9):2007-2016.
[12] CHAE E.Frequency domain analysis of discrete wavelet transform coefficients and Their adaptive shrinkage for anti-aliasing[J].IEEE International Conference on Image Processing,Melbourne,Australia,2013,10:1071-1074.
[13] KIM.P contrast enhancement using brightness preserving bi-histogram equalization[J].Consumer Electronics IEEE Transactions on,1997,43(1):1-8.
[14] BAI X Z,ZHOU F G,XUE B D.Image enhance-ment using multi scale image features extracted by top-hat transform[J].Optics&Laser Technology.2014,44(2):328-336.
[15] 袁云梅,多化琼,马坤.基于权重系数的木材图像增强与识别[J].西北林学院学报,2018,33(2):209-212.YUAN Y M,DUO H Q,MA K.Wood image enhancement and recognition based on weight coefficient[J].Journal of Northwest Forestry University,2018,33(2):209-212.(in Chinese)
[16] 多化豫,孙枭雄,袁云梅.基于图像处理提高木材识别准确性的新方法[J].西北林学院学报,2017,32(1):244-247.DUO H Y,SUN X X.YUAN Y M.Improvement of wood identification accuracy based on image processing[J].Journal of Northwest Forestry University,2017,32(1):244-247.(in Chinese)
[2] 安静,张贵仓,刘燕妮.基于多尺度top-hat变换的自适应彩色图像增强[J].计算机工程与科学,2017,3(7):1317-1321.AN J,ZHANG G C,LIU Y N.Adaptive color image enhancement based on multi-scale top-hat transform[J].Computer Engineering&Science,2017,3(7):1317-1321.(in Chinese)
[3] 杨艳,陈章宝,王艳春.基于四元数傅里叶变换的彩色图像增强[J].蚌埠学院学报,2017,6(6):9-12.YANG Y,CHEN Z B,WANG Y C.Color images enhancing based on quaternion fourier transforms[J].Journal of Benghu University,2017,6(6):9-12.(in Chinese)
[4] 何惜琴,许艳华.一种改进的仿生彩色图像增强算法[J].闽江学院学报,2017(5):49-55.HENG X Q,XU Y H.An improved bio-inspired algorithm for color image enhancement[J].Journal of Minjiang University,2017(5):49-55.(in Chinese)
[5] JOBSON D J,RAHMA N Z,WOODLL G A.Properties and perform-ance of a center/surround retinex[J].IEEE Transactions on Image Processing,1997,6(3):451-462.
[6] RAHMAN Z U,JOBSON D J,WOODELL G A.Retinex processing for automatic images enhancement[J].Journal of Electrontic Imaging,2004,13(1):100-110.
[7] 蔡式东,杨芳.基于直方图修正的图像增强算法[J].光电子技术,2016.9,32(3):155-160.CAI S D,YANG F.Image enhancement based on histogram modification[J].Opto Electronic Technology,2016,9,32(3):155-160.(in Chinese)
[8] 韩磊,曲中水.一种RGB模型彩色图像增强方法[J].哈尔滨理工大学学报,2014,19(6):59-64.HAN L,QU Z S.A kind of enhancement method of color image based on RGB color model[J].Journal of Harbin University of Science And Technology,2014,19(6):59-64.(in Chinese)
[9] GAN J.Non-separable wavelet-like lifting structure for image and video compression with aliasing suppression[J].IEEE International Conference on Image Processing,San Antonio,USA,2007.10(6):65-68.
[10] 杜颖财,宋路,万秋华.小波变换实现的光电编码器精确实时测速[J].红外与激光工程,2017,46(5):1-6.DU Y C,SONG L,WAN Q H.Accurate real-time velocity measurement of photoelectric encoder based on wavelet transform[J].Infrared and Laser Engineering,2017,46(5):1-6.(in Chinese)
[11] COULANGE B.An aliasing detection algorithm based on suspicious colocalizations of fourier coefficients[J].IEEE International Conference on Image Processing,HongKong,China,2010(9):2007-2016.
[12] CHAE E.Frequency domain analysis of discrete wavelet transform coefficients and Their adaptive shrinkage for anti-aliasing[J].IEEE International Conference on Image Processing,Melbourne,Australia,2013,10:1071-1074.
[13] KIM.P contrast enhancement using brightness preserving bi-histogram equalization[J].Consumer Electronics IEEE Transactions on,1997,43(1):1-8.
[14] BAI X Z,ZHOU F G,XUE B D.Image enhance-ment using multi scale image features extracted by top-hat transform[J].Optics&Laser Technology.2014,44(2):328-336.
[15] 袁云梅,多化琼,马坤.基于权重系数的木材图像增强与识别[J].西北林学院学报,2018,33(2):209-212.YUAN Y M,DUO H Q,MA K.Wood image enhancement and recognition based on weight coefficient[J].Journal of Northwest Forestry University,2018,33(2):209-212.(in Chinese)
[16] 多化豫,孙枭雄,袁云梅.基于图像处理提高木材识别准确性的新方法[J].西北林学院学报,2017,32(1):244-247.DUO H Y,SUN X X.YUAN Y M.Improvement of wood identification accuracy based on image processing[J].Journal of Northwest Forestry University,2017,32(1):244-247.(in Chinese)