小波分析在木材缺陷图像处理中的应用
|
摘要
随着木材科学的发展,采用X射线作为无损检测手段,应用计算机数字图像处理技术对木材缺陷图像进行处理已成木材无损检测研究的一个重要方向。应用木材X射线无损检测系统获取木材缺陷图像的过程中,由于摄像系统质量、给光条件以及录入装置等因素的影响,使木材缺陷图像增加了噪声,恶化了图像质量,图像变模糊,特征被淹没,对比度降低,给木材缺陷的识别带来困难。小波分析作为当前一种流行的数字图像处理工具,其良好的时—频局部化特点和多尺度的特性使其在图像处理中得到了广泛的应用。针对木材缺陷图像的特点,将小波分析应用到图像去噪和图像边缘检测中,使图像更加清晰,易于识别。
本文对小波分析的基本理论、多分辨分析、MALLAT算法、小波基选取等一系列相关内容进行了分析。重点研究了运用小波变换对木材缺陷图像进行去噪和边缘检测的原理和具体实现方法,并与传统的图像去噪方法和经典的边缘检测方法进行了比较。
木材缺陷图像的小波去噪方法中,重点研究了小波阈值去噪。对去噪小波、阈值、阈值函数的选取几个关键问题进行了详细分析与讨论。运用全局阂值、局部阈值两种不同阈值；硬阈值、软阈值和高频系数置零三种不同的阈值化函数对木材缺陷图像进行小波去噪处理。以均方误差(MSE)和峰值信噪比(PSNR)作为评价标准,将小波阈值去噪方法与传统的图像去噪方法作仿真对比实验,得出适合木材缺陷图像的去噪方法。单一的去噪方法不能满足图像处理的需要,提出小波包与数学形态学结合对木材缺陷图像去噪的方法,将二者的优点有效地结合起来,得到较高峰值信噪比,改善了主观视觉效果,优于单一的去噪方法。
针对单一尺度的边缘检测不利于对图像细节定位和提取的缺点,小波变换的多尺度特性为边缘检测提供一种新方法。本文选取B样条小波作为边缘检测小波,应用基于二进小波变换的多尺度边缘检测算法实现对木材缺陷图像的边缘检测。除此之外,小波包具有对低频、高频部分进行分解,获得更多的图像信息的特性,使重构后得到的近似部分图像去除高频分量,检测到更加清晰、连续的边缘,将小波包用于木材缺陷图像的边缘检测中,也得到较好的实验效果。
With the development of wood science, using X-ray as non-destructive testing method and computer digital image processing technology for wood defect images became an important direction of wood non-destructive testing. During the wood defect images were obtained by wood X-ray non-destructive testing system, factors such as the camera system quality, the given light condition and input device increased noise to deteriorate image quality, make image fuzzy, submerge features and reduce contrast. It was different to identify wood defect.As a popular digital image processing tool, wavelet analysis had been widely used in image processing because of its good time-frequency localization characteristic and multi-scale feature. According to the characteristics of wood defect images, wavelet analysis had been used in image denoising and image edge detection to make image clearer, easy to identify.
The basic theories of wavelet analysis, multi-resolution analysis, MALLAT algorithm, wavelet base selection and other related contents were analysed in this paper. The principle and realization methods of image denoising and edge detection algorithm using wavelet transform were researched, compared with traditional image denoising methods and classic edge detection methods.
The wavelet threshold denoising was the key research among wavelet denoising methods for wood defect images. Several key problems of denoising wavelet, threshold and threshold function selection had been carried on analysis and discussion in detail. Two different thresholds of global threshold and local threshold, three different threshold value initialization functions of hard threshold, soft threshold and high frequency coefficients set zero were used to denoise processing for wood defect images. With Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) as evaluation standard, wavelet threshold denoising algorithm and the traditional image denoising algorithm had been done simulation experiments to find the suitable denoising method for wood defect image. Single denoising method could not meet the need of image processing. Denoising method based on wavelet packet combined with gray-scale morphological filtering was proposed for wood defect image. The method received better PSNR and visual effect. It was better than other denoising methods.
Single scale edge detection was not good for image detail position and extraction. Multi-scale characteristic of wavelet transform provided a new method for edge detection. B-spline wavelet as edge detection wavelet, application of multi-scale edge detection algorithm based on the dyadic wavelet transforms achieved edge detection for wood defect images.In addition, wavelet packet can decompose low frequency and high frequency to obtain more image information. The approximate part which was obtained by reconstruction can remove the high frequency to detect clearer, more continuous edge. Wavelet packet was used in edge detection for wood defect images and also got good results.
本文对小波分析的基本理论、多分辨分析、MALLAT算法、小波基选取等一系列相关内容进行了分析。重点研究了运用小波变换对木材缺陷图像进行去噪和边缘检测的原理和具体实现方法,并与传统的图像去噪方法和经典的边缘检测方法进行了比较。
木材缺陷图像的小波去噪方法中,重点研究了小波阈值去噪。对去噪小波、阈值、阈值函数的选取几个关键问题进行了详细分析与讨论。运用全局阂值、局部阈值两种不同阈值；硬阈值、软阈值和高频系数置零三种不同的阈值化函数对木材缺陷图像进行小波去噪处理。以均方误差(MSE)和峰值信噪比(PSNR)作为评价标准,将小波阈值去噪方法与传统的图像去噪方法作仿真对比实验,得出适合木材缺陷图像的去噪方法。单一的去噪方法不能满足图像处理的需要,提出小波包与数学形态学结合对木材缺陷图像去噪的方法,将二者的优点有效地结合起来,得到较高峰值信噪比,改善了主观视觉效果,优于单一的去噪方法。
针对单一尺度的边缘检测不利于对图像细节定位和提取的缺点,小波变换的多尺度特性为边缘检测提供一种新方法。本文选取B样条小波作为边缘检测小波,应用基于二进小波变换的多尺度边缘检测算法实现对木材缺陷图像的边缘检测。除此之外,小波包具有对低频、高频部分进行分解,获得更多的图像信息的特性,使重构后得到的近似部分图像去除高频分量,检测到更加清晰、连续的边缘,将小波包用于木材缺陷图像的边缘检测中,也得到较好的实验效果。
With the development of wood science, using X-ray as non-destructive testing method and computer digital image processing technology for wood defect images became an important direction of wood non-destructive testing. During the wood defect images were obtained by wood X-ray non-destructive testing system, factors such as the camera system quality, the given light condition and input device increased noise to deteriorate image quality, make image fuzzy, submerge features and reduce contrast. It was different to identify wood defect.As a popular digital image processing tool, wavelet analysis had been widely used in image processing because of its good time-frequency localization characteristic and multi-scale feature. According to the characteristics of wood defect images, wavelet analysis had been used in image denoising and image edge detection to make image clearer, easy to identify.
The basic theories of wavelet analysis, multi-resolution analysis, MALLAT algorithm, wavelet base selection and other related contents were analysed in this paper. The principle and realization methods of image denoising and edge detection algorithm using wavelet transform were researched, compared with traditional image denoising methods and classic edge detection methods.
The wavelet threshold denoising was the key research among wavelet denoising methods for wood defect images. Several key problems of denoising wavelet, threshold and threshold function selection had been carried on analysis and discussion in detail. Two different thresholds of global threshold and local threshold, three different threshold value initialization functions of hard threshold, soft threshold and high frequency coefficients set zero were used to denoise processing for wood defect images. With Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) as evaluation standard, wavelet threshold denoising algorithm and the traditional image denoising algorithm had been done simulation experiments to find the suitable denoising method for wood defect image. Single denoising method could not meet the need of image processing. Denoising method based on wavelet packet combined with gray-scale morphological filtering was proposed for wood defect image. The method received better PSNR and visual effect. It was better than other denoising methods.
Single scale edge detection was not good for image detail position and extraction. Multi-scale characteristic of wavelet transform provided a new method for edge detection. B-spline wavelet as edge detection wavelet, application of multi-scale edge detection algorithm based on the dyadic wavelet transforms achieved edge detection for wood defect images.In addition, wavelet packet can decompose low frequency and high frequency to obtain more image information. The approximate part which was obtained by reconstruction can remove the high frequency to detect clearer, more continuous edge. Wavelet packet was used in edge detection for wood defect images and also got good results.
引文
[1]李坚.木材科学[M].哈尔滨：东北林业大学出版社,1999：108～121
[2]李家伟,陈积懋.无损检测手册[M].北京：机械工业出版社,2002：392～397
[3]张磊.有限冲激相应数字滤波器在电磁无损检测系统中的应用[D].哈尔滨理工大学硕士论文.2008：1～4
[4]崔金刚.图像处理技术在木材检测方面的应用[D].东北林业大学硕士论文.2007：1～4
[5]梁善庆,蔡智勇,王喜平等.北美木材无损检测技术的研究与应用[J].木材工业,2008,22(3)：5～8
[6]王立海,杨学春,徐凯宏.木材无损检测技术的研究现状与进展[J].森林工程,2001,17(6)：1～3
[7]姜忠华,申世杰,翟志文.木材无损检测的现状与发展趋势[J].林业机械与木工设备,2010,38(2)：4～7
[8]J. Oja, S. Grundberg, A. Gronlund. Predicting the Strength of Sawn Products by X-ray scanning of Logs [J]. A preliminary study. Wood and Fiber Science,2000,32 (2):203～208
[9]G. S. Schajer. Lumber Strength Grading using X-ray Scanning [J]. Forest Products Journal, 2001,51(1):43～50
[10]李慧,撒潮,李丽.单板密度β射线检测法的研究[J].林业机械与木工设备,2007,35(8)：19～22
[11]王新,申世杰.木材无损检测研究概况与发展趋势[J].东北林业大学学报,2009,31(1)：202～205
[12]连广,王廷魁,戴澄月.X射线强度直接计数法无损检测木材生长轮密度[J].东北林业大学学报,1990,18(5)：38～42
[13]冯星球,陈焕娣,张立芳.用X射线研究人造板材的密度梯度[J].南京林业大学学报,1995,19(4)：62～66
[14]江泽慧,黄安民,费本华等.利用近红外光谱和X射线衍射技术分析木材微纤丝角[J].光谱学与光谱分析,2006,26(7)：1230～1233
[15]戚大伟,牟洪波.人工神经网络在木材缺陷检测中的应用[J].森林工程,2006,22(1)：21～23
[16]李永生,戚大伟.基于FCM的木材X射线图像缺陷分割[J].森林工程,2008,24(5)：51～52
[17]林剑,赵广杰,孟令萱等.利用X射线衍射技术与红外光谱分析真菌侵蚀的木材[J].光谱学与光谱分析,2010,30(6)：1674～1677
[18]刘自强,乔景绿,戚大伟.木材缺陷图象的伪彩色处理[J].林业科学,1989,(02)：185～189
[19]孙丽萍,张汝楠,秦怀光.X射线在原木无损检测中的应用[J].林业机械与木工设备,2007,35(10)：54～55
[20]M. Paggi, A.Carpinteri. Fractal and Multifraetal Approaches for the Analysis of Crack-Size Dependent Scaling Laws in Fatigue [J]. Chaos, Solitons & Fractals,2009,40(3):1136-1145
[21]E. Aptoula, S. Lefevre. A Comparative Study on Multivariate Mathematical Morphology [J]. Pattern Recognition,2007,40(11):2914～2929
[22]J.GM. Schavemaker, M.J.T. Reinders, J.J. Gerbrands. Image Sharpening by Morphological Filtering [J]. Pattern Recognition,2000,33(6):997～1012
[23]李学益,戚大伟.木材缺陷图像处理软件的研制[J].东北林业大学学报,2002,30(03)：79～80
[24]D. Feng, K. Lau. Multifraetal Formalism for Self-similar Measures with Weak Separation Condition [J]. Journal de Mathematiques Pures et Applique,2009,92(4):407～428
[25]R. Lopes, N. Betrouni. Fractal and Multifractal Analysis:A review [J]. Medical Image Analysis,2009,13(4):634～649
[26]张军团,林君,李绚.基于分形维数的SAR图像变化检测[J].西安工业大学学报,2008,28(05)：466～470
[27]P. Shang. Multifractal Characteristics of Palm Print and its Extracted Algorithm [J]. Applied Mathematical Modelling,2009,3(12):4378～4387
[28]D.L. Schmoldt, J. He, A.L. Abbott. Automated Labeling of Log Features in CT Imagery of Multiple Hardwood Species [J]. Wood and Fiber Science,2000,32(3):287～300
[29]丁建文,王宝金,徐兆军.基于灰度直方图分析的原木CT图象分割[J].木材加工机械,2006,(5)：1～4
[30]D.L. Schmoldt, P. Li, A.L. Abbott. Machine Vision Using Artificial Neural Networks with Local 3D Neighborhoods [J]. Computers and Electronics in Agriculture,1997,16(3):255～271
[31]陈书海,傅录祥.实用数字图像处理[M].北京：科学出版社,2005：1～7
[32]刘伟元,张铁军,戴亮宇等.小波分析在超声波探测数据处理中的应用[J].湖南交通科技,2006,32(4)：140～142
[33]孙延奎.小波分析及其应用[M].北京：机械工业出版社,2006：171～175
[34]Z.R. Struzik. Wavelet Methods in Time-series Processing [J]. Physics A:Statistical Mechanics and its Application,2001,296(1～2):307～319
[35]韩小燕,孟广耀,王威.小波技术在逆向工程中的应用[J].现代制造技术与装备,2009,(6)：4～6
[36]张一,成礼智.小波变换图像压缩中最优小波基的选取方法[J].电视技术,2004,(10)：4～7
[37]I. Atkinson, F. Kamalabadi. Basis Selection for Wavelet Processing of Sparse Source Signals [C]. IEEE International Conference On Acoustics, Speech And Signal Processing,2007,3:15～20
[38]B. Benjamin, B. Georgic. Entropy-Threshold Method for Best Basis Selection [J]. Image and Vision Computing,2001,19:477～484
[39]赵学智,林颖,陈文戈等.奇异性信号检测时小波基的选择[J].华南理工大学学报,2000,28(10)：75～80
[40]戚大伟.基于分形理论的原木缺陷X射线图像分析与处理[D].东北林业大学博士论文. 2003：2～7
[41]李旭超,朱善安.小波域图像降噪概述[J].中国图象图形学报,2006,11(9)：1201～1208
[42]龙振弘.数字图像去噪典型算法的分析比较[J].福建广播电视大学学报,2009,(1)：67～69
[43]艾泽潭,石庚辰.小波变换在图像去噪中的应用[J].科技导报,2010,28(1)：102～106
[44]何瑞强.基于小波变换的图像去噪方法[D].西安建筑科技大学硕士论文.2009：28～30
[45]沈飞.小波变换在图像处理中的应用[D].哈尔滨工程大学硕士论文.2009：22～36
[46]罗军辉,冯平,哈力旦.MATLAB 7.0在图像处理中的应用[M].北京：机械工业出版社,2005：135～142
[47]贺兴华,周媛媛,王继阳. MATLAB 7.X图像处理[M].北京：人民邮电出版社,2006：95～96
[48]杨秀君.基于图像处理的罐装液位检测方法的研究[D].湖南大学硕士论文.2009：21～26
[49]D. Tu, J.J. Shen. The Design of Wavelet Domain Wiener Filter and its Application in Image Denoising [J]. Systems Engineering and Electronics,2001,23:4～7
[50]聂洪山,沈振康.一种基于Wiener滤波的红外背景抑制方法[J].国防科技大学学报,2003,25(3)：54～57
[51]王浩,曾欢.几种图像去噪平滑算法的性能分析[J].科技咨讯,2006(3)：1～2
[52]柏春岚.基于频域的图像增强及其MATLAB实现[J].测绘,2009,32(6)：273～275
[53]成礼智,王红霞,罗永.小波的理论与应用[M].北京：科学出版社,2004：269～270
[54]Q. Pan, R. Zhang, G. Z. Dai, Two Denoising Methods by Wavelet Transforms [J]. IEEE Transactions on Signal Processing,1999,47(12):3401-3406
[55]H.C. Zhao, et al. Efficient Wavelet-based Image Denoising Algorithm [J]. IEEE Trans Electronics Letters,2001(37):683～685
[56]李月琴,栗苹,闫晓鹏等.小波变换模极大去噪法在无线电引信信号处理中的应用[J].兵工学报,2008,29(10)：1172～1176
[57]林椹甚少,宋国乡,薛文.图像的几种小波去噪方法的比较与改进[J].西安电子科技大学学报,2004,(31)4：626～629
[58]张翠芳.小波降噪及其在MATLAB中的仿真与比较[J].信息技术,2008,(3)：4～5
[59]刘钰,马艳丽,刘艳霞.小波阈值图像去噪算法及MATLAB仿真实验[J].数字技术与应用,2010,(6)：122～123
[60]陈婷,郭金翠,黄文丽.图像去噪的小波阈值法研究[J].软件导刊,2010,9(2)：166～167
[61]K. Berkner, R. O. Wells. Smoothness Estimates for Soft-threshold Denoising via Translation-invariant Wavelet Transforms. Applied and Computational Harmonic Analysis,2002,12(1):1～24
[62]S. Sudha, G.R. Surssh, R. Sukanesh. Wavelet based Image Denoising using Adaptive Thresholding [C]. International Conference on Computational Intelligence and Multimedia Applications,2007:296～300
[63]Y. Chen, C. Han. Adaptive Wavelet Threshold for Image De-nosing [J]. IEEE Electronics Letters,2005,10(11)
[64]王静,范毅方.基于小波阈值的SEMG信号消噪方法研究[J].计算机仿真,2010,27(2)：325～327
[65]飞思科技产品研发中心.小波分析理论与MATLAB 7实现[M].北京：电子工业出版社,2005：321～330
[66]D.L. Donoho. Denoising by Soft-thresholding [J]. IEEE Trans.Inform.Theory,1995,41:613-627
[67]S.G. Chang, B. Yu, M. Vetterli. Adaptive Wavelet Thresholding for Image Denoising and Compression [J]. IEEE Trans on Image Proeessing,2000,9(9):1532～1546
[68]Y. S. Dong, X.M. Yi. Wavelet Denoising based on four Modified Functions for Threshold Estimation [J]. Journal of Mathematics,2005.
[69]Y.H. Peng. De-noising by Modified Soft-Thresholding [J]. IEEE Trans on IT,2000,63:760-762
[70]孙崟培,王朝英.小波分析和小波包在图像消噪中的应用[J].通信技术,2009,42(1)：285～287
[71]C.R. Giarding, E.R. Doughterty. Morphological Methods in Image and Signal Processing [M]. NewJersery:Prentice Hall,1988:5～8
[72]李慧,蔺启忠,王钦军等.基于小波包变换和数学形态学结合的光谱去噪方法研究[J].光谱学与光谱分析,2010,30(3)：644～648
[73]程正兴.小波分析与应用实例[M].西安：西安交通大学出版社,2006：214～215
[74]陈颉,陈启祥.数字图像的边缘检测[J].湖北工业大学学报,2006,21(4)：108～111
[75]张伟.小波分析在图像边缘检测中的应用[J].无损探伤,2008,32(4)：8～9
[76]T. Loupas, W.N. Mcdichen. An Adaptive Weighted Median Filter for Speckle Suppression in Medical Ultrasonic Images [J]. IEEE Trans on Circuits Cyst,2002,36(1):129～135
[77]D. Marr, E. Hildreth. Theory of Edge Detection [J]. Proc.Roy.Soc,2008,27:187～217
[78]C. Gonzaez, E. Woods. Digital Image Processing [J]. Second Edition. House of Electronice Industry,2003:463～474
[79]张德丰.图像多尺度边缘检测算法研究[J].计算机工程与设计,2008,29(21)：5524～5525
[80]罗辉,张铁中,杨丽.小波变换在农田图像边缘检测中的应用[J].农机化研究,2006,(2)：157～160
[81]舒小华,刘耦耕.小波变换在医学图像边缘提取中的应用[J].现代电子技术,2005,8：66～67
[82]M. Tabb, N. Ahuja. Multiscale Image Segmentation by Integrated Edge and Region [J]. IEEE Trans on Image Processing,1997,5(6):642～655
[83]J.M. Niya, A. Aghagolzadeh. Edge Detection using Directional Wavelet Transforms [C]. Proceedings of the 12th IEEE Mediterranean Electrotechnical Conference,2004:281～284
[84]E. Brannock, M. Weeks. Edge Detection using Wavelets [C].2006 Proceedings of the 44th annual Southeast Regional Conference,2006:649～654
[85]刘飞飞,董绍华.基于小波多尺度分析的图像边缘检测[J].微计算机信息,2009,25(3)：271～272
[86]B. Tremblais, B. Augereau. A fast multi-scale edge detection algorithm [J]. Pattern Recognition letters,2004,25:603～618
[87]于彤,马社祥,徐华.基于二进小波变换的图像边缘检测算法研究[J].天津理大学学报,2009,25(1)：59～62
[88]王兰.基于小波变换的图像边缘检测算法的研究[D].武汉科技大学硕士论文.2009：27～31
[89]YP. Wang, L. Lees. Scale-space derived from B-splines [J]. IEEE Trans. PAMI,1998,20(10): 1040～1055
[90]梁艳.基于B样条小波变换的图像边缘检测技术研究[D].武汉理工大学硕士论文.2008：37～50
[91]Q. Jiang. Orthogonal Multiwavelets with Optimum Time-frequency Resolution [J]. IEEE Trans. SP. Special Issue on Theory and Application of Filterbanks and Wavelet Transforms,1998,46(4): 830～844
[92]S. Mallat. Multiresolution Approximation and Wavelet Orthonormal Bases of L2(R) [J]. Trans. Amer. Math. Soc,1959,315(8):59～67
[93]S. Mattlat. A theory for Multiresolution Sidecomosition:the Wavelet Representation [J]. IEEE Trans. PAMI,1989,11(7):674～693
[94]S. Mallat. Multifrequeney Channel Decomposition of Images and Wavelets [J]. IEEE Trans. ASSP,1989,37(12):2091～2110
[95]刘建军.基于小波包的边缘检测算法研究[D].中南大学硕士论文.2010：44～52
[2]李家伟,陈积懋.无损检测手册[M].北京：机械工业出版社,2002：392～397
[3]张磊.有限冲激相应数字滤波器在电磁无损检测系统中的应用[D].哈尔滨理工大学硕士论文.2008：1～4
[4]崔金刚.图像处理技术在木材检测方面的应用[D].东北林业大学硕士论文.2007：1～4
[5]梁善庆,蔡智勇,王喜平等.北美木材无损检测技术的研究与应用[J].木材工业,2008,22(3)：5～8
[6]王立海,杨学春,徐凯宏.木材无损检测技术的研究现状与进展[J].森林工程,2001,17(6)：1～3
[7]姜忠华,申世杰,翟志文.木材无损检测的现状与发展趋势[J].林业机械与木工设备,2010,38(2)：4～7
[8]J. Oja, S. Grundberg, A. Gronlund. Predicting the Strength of Sawn Products by X-ray scanning of Logs [J]. A preliminary study. Wood and Fiber Science,2000,32 (2):203～208
[9]G. S. Schajer. Lumber Strength Grading using X-ray Scanning [J]. Forest Products Journal, 2001,51(1):43～50
[10]李慧,撒潮,李丽.单板密度β射线检测法的研究[J].林业机械与木工设备,2007,35(8)：19～22
[11]王新,申世杰.木材无损检测研究概况与发展趋势[J].东北林业大学学报,2009,31(1)：202～205
[12]连广,王廷魁,戴澄月.X射线强度直接计数法无损检测木材生长轮密度[J].东北林业大学学报,1990,18(5)：38～42
[13]冯星球,陈焕娣,张立芳.用X射线研究人造板材的密度梯度[J].南京林业大学学报,1995,19(4)：62～66
[14]江泽慧,黄安民,费本华等.利用近红外光谱和X射线衍射技术分析木材微纤丝角[J].光谱学与光谱分析,2006,26(7)：1230～1233
[15]戚大伟,牟洪波.人工神经网络在木材缺陷检测中的应用[J].森林工程,2006,22(1)：21～23
[16]李永生,戚大伟.基于FCM的木材X射线图像缺陷分割[J].森林工程,2008,24(5)：51～52
[17]林剑,赵广杰,孟令萱等.利用X射线衍射技术与红外光谱分析真菌侵蚀的木材[J].光谱学与光谱分析,2010,30(6)：1674～1677
[18]刘自强,乔景绿,戚大伟.木材缺陷图象的伪彩色处理[J].林业科学,1989,(02)：185～189
[19]孙丽萍,张汝楠,秦怀光.X射线在原木无损检测中的应用[J].林业机械与木工设备,2007,35(10)：54～55
[20]M. Paggi, A.Carpinteri. Fractal and Multifraetal Approaches for the Analysis of Crack-Size Dependent Scaling Laws in Fatigue [J]. Chaos, Solitons & Fractals,2009,40(3):1136-1145
[21]E. Aptoula, S. Lefevre. A Comparative Study on Multivariate Mathematical Morphology [J]. Pattern Recognition,2007,40(11):2914～2929
[22]J.GM. Schavemaker, M.J.T. Reinders, J.J. Gerbrands. Image Sharpening by Morphological Filtering [J]. Pattern Recognition,2000,33(6):997～1012
[23]李学益,戚大伟.木材缺陷图像处理软件的研制[J].东北林业大学学报,2002,30(03)：79～80
[24]D. Feng, K. Lau. Multifraetal Formalism for Self-similar Measures with Weak Separation Condition [J]. Journal de Mathematiques Pures et Applique,2009,92(4):407～428
[25]R. Lopes, N. Betrouni. Fractal and Multifractal Analysis:A review [J]. Medical Image Analysis,2009,13(4):634～649
[26]张军团,林君,李绚.基于分形维数的SAR图像变化检测[J].西安工业大学学报,2008,28(05)：466～470
[27]P. Shang. Multifractal Characteristics of Palm Print and its Extracted Algorithm [J]. Applied Mathematical Modelling,2009,3(12):4378～4387
[28]D.L. Schmoldt, J. He, A.L. Abbott. Automated Labeling of Log Features in CT Imagery of Multiple Hardwood Species [J]. Wood and Fiber Science,2000,32(3):287～300
[29]丁建文,王宝金,徐兆军.基于灰度直方图分析的原木CT图象分割[J].木材加工机械,2006,(5)：1～4
[30]D.L. Schmoldt, P. Li, A.L. Abbott. Machine Vision Using Artificial Neural Networks with Local 3D Neighborhoods [J]. Computers and Electronics in Agriculture,1997,16(3):255～271
[31]陈书海,傅录祥.实用数字图像处理[M].北京：科学出版社,2005：1～7
[32]刘伟元,张铁军,戴亮宇等.小波分析在超声波探测数据处理中的应用[J].湖南交通科技,2006,32(4)：140～142
[33]孙延奎.小波分析及其应用[M].北京：机械工业出版社,2006：171～175
[34]Z.R. Struzik. Wavelet Methods in Time-series Processing [J]. Physics A:Statistical Mechanics and its Application,2001,296(1～2):307～319
[35]韩小燕,孟广耀,王威.小波技术在逆向工程中的应用[J].现代制造技术与装备,2009,(6)：4～6
[36]张一,成礼智.小波变换图像压缩中最优小波基的选取方法[J].电视技术,2004,(10)：4～7
[37]I. Atkinson, F. Kamalabadi. Basis Selection for Wavelet Processing of Sparse Source Signals [C]. IEEE International Conference On Acoustics, Speech And Signal Processing,2007,3:15～20
[38]B. Benjamin, B. Georgic. Entropy-Threshold Method for Best Basis Selection [J]. Image and Vision Computing,2001,19:477～484
[39]赵学智,林颖,陈文戈等.奇异性信号检测时小波基的选择[J].华南理工大学学报,2000,28(10)：75～80
[40]戚大伟.基于分形理论的原木缺陷X射线图像分析与处理[D].东北林业大学博士论文. 2003：2～7
[41]李旭超,朱善安.小波域图像降噪概述[J].中国图象图形学报,2006,11(9)：1201～1208
[42]龙振弘.数字图像去噪典型算法的分析比较[J].福建广播电视大学学报,2009,(1)：67～69
[43]艾泽潭,石庚辰.小波变换在图像去噪中的应用[J].科技导报,2010,28(1)：102～106
[44]何瑞强.基于小波变换的图像去噪方法[D].西安建筑科技大学硕士论文.2009：28～30
[45]沈飞.小波变换在图像处理中的应用[D].哈尔滨工程大学硕士论文.2009：22～36
[46]罗军辉,冯平,哈力旦.MATLAB 7.0在图像处理中的应用[M].北京：机械工业出版社,2005：135～142
[47]贺兴华,周媛媛,王继阳. MATLAB 7.X图像处理[M].北京：人民邮电出版社,2006：95～96
[48]杨秀君.基于图像处理的罐装液位检测方法的研究[D].湖南大学硕士论文.2009：21～26
[49]D. Tu, J.J. Shen. The Design of Wavelet Domain Wiener Filter and its Application in Image Denoising [J]. Systems Engineering and Electronics,2001,23:4～7
[50]聂洪山,沈振康.一种基于Wiener滤波的红外背景抑制方法[J].国防科技大学学报,2003,25(3)：54～57
[51]王浩,曾欢.几种图像去噪平滑算法的性能分析[J].科技咨讯,2006(3)：1～2
[52]柏春岚.基于频域的图像增强及其MATLAB实现[J].测绘,2009,32(6)：273～275
[53]成礼智,王红霞,罗永.小波的理论与应用[M].北京：科学出版社,2004：269～270
[54]Q. Pan, R. Zhang, G. Z. Dai, Two Denoising Methods by Wavelet Transforms [J]. IEEE Transactions on Signal Processing,1999,47(12):3401-3406
[55]H.C. Zhao, et al. Efficient Wavelet-based Image Denoising Algorithm [J]. IEEE Trans Electronics Letters,2001(37):683～685
[56]李月琴,栗苹,闫晓鹏等.小波变换模极大去噪法在无线电引信信号处理中的应用[J].兵工学报,2008,29(10)：1172～1176
[57]林椹甚少,宋国乡,薛文.图像的几种小波去噪方法的比较与改进[J].西安电子科技大学学报,2004,(31)4：626～629
[58]张翠芳.小波降噪及其在MATLAB中的仿真与比较[J].信息技术,2008,(3)：4～5
[59]刘钰,马艳丽,刘艳霞.小波阈值图像去噪算法及MATLAB仿真实验[J].数字技术与应用,2010,(6)：122～123
[60]陈婷,郭金翠,黄文丽.图像去噪的小波阈值法研究[J].软件导刊,2010,9(2)：166～167
[61]K. Berkner, R. O. Wells. Smoothness Estimates for Soft-threshold Denoising via Translation-invariant Wavelet Transforms. Applied and Computational Harmonic Analysis,2002,12(1):1～24
[62]S. Sudha, G.R. Surssh, R. Sukanesh. Wavelet based Image Denoising using Adaptive Thresholding [C]. International Conference on Computational Intelligence and Multimedia Applications,2007:296～300
[63]Y. Chen, C. Han. Adaptive Wavelet Threshold for Image De-nosing [J]. IEEE Electronics Letters,2005,10(11)
[64]王静,范毅方.基于小波阈值的SEMG信号消噪方法研究[J].计算机仿真,2010,27(2)：325～327
[65]飞思科技产品研发中心.小波分析理论与MATLAB 7实现[M].北京：电子工业出版社,2005：321～330
[66]D.L. Donoho. Denoising by Soft-thresholding [J]. IEEE Trans.Inform.Theory,1995,41:613-627
[67]S.G. Chang, B. Yu, M. Vetterli. Adaptive Wavelet Thresholding for Image Denoising and Compression [J]. IEEE Trans on Image Proeessing,2000,9(9):1532～1546
[68]Y. S. Dong, X.M. Yi. Wavelet Denoising based on four Modified Functions for Threshold Estimation [J]. Journal of Mathematics,2005.
[69]Y.H. Peng. De-noising by Modified Soft-Thresholding [J]. IEEE Trans on IT,2000,63:760-762
[70]孙崟培,王朝英.小波分析和小波包在图像消噪中的应用[J].通信技术,2009,42(1)：285～287
[71]C.R. Giarding, E.R. Doughterty. Morphological Methods in Image and Signal Processing [M]. NewJersery:Prentice Hall,1988:5～8
[72]李慧,蔺启忠,王钦军等.基于小波包变换和数学形态学结合的光谱去噪方法研究[J].光谱学与光谱分析,2010,30(3)：644～648
[73]程正兴.小波分析与应用实例[M].西安：西安交通大学出版社,2006：214～215
[74]陈颉,陈启祥.数字图像的边缘检测[J].湖北工业大学学报,2006,21(4)：108～111
[75]张伟.小波分析在图像边缘检测中的应用[J].无损探伤,2008,32(4)：8～9
[76]T. Loupas, W.N. Mcdichen. An Adaptive Weighted Median Filter for Speckle Suppression in Medical Ultrasonic Images [J]. IEEE Trans on Circuits Cyst,2002,36(1):129～135
[77]D. Marr, E. Hildreth. Theory of Edge Detection [J]. Proc.Roy.Soc,2008,27:187～217
[78]C. Gonzaez, E. Woods. Digital Image Processing [J]. Second Edition. House of Electronice Industry,2003:463～474
[79]张德丰.图像多尺度边缘检测算法研究[J].计算机工程与设计,2008,29(21)：5524～5525
[80]罗辉,张铁中,杨丽.小波变换在农田图像边缘检测中的应用[J].农机化研究,2006,(2)：157～160
[81]舒小华,刘耦耕.小波变换在医学图像边缘提取中的应用[J].现代电子技术,2005,8：66～67
[82]M. Tabb, N. Ahuja. Multiscale Image Segmentation by Integrated Edge and Region [J]. IEEE Trans on Image Processing,1997,5(6):642～655
[83]J.M. Niya, A. Aghagolzadeh. Edge Detection using Directional Wavelet Transforms [C]. Proceedings of the 12th IEEE Mediterranean Electrotechnical Conference,2004:281～284
[84]E. Brannock, M. Weeks. Edge Detection using Wavelets [C].2006 Proceedings of the 44th annual Southeast Regional Conference,2006:649～654
[85]刘飞飞,董绍华.基于小波多尺度分析的图像边缘检测[J].微计算机信息,2009,25(3)：271～272
[86]B. Tremblais, B. Augereau. A fast multi-scale edge detection algorithm [J]. Pattern Recognition letters,2004,25:603～618
[87]于彤,马社祥,徐华.基于二进小波变换的图像边缘检测算法研究[J].天津理大学学报,2009,25(1)：59～62
[88]王兰.基于小波变换的图像边缘检测算法的研究[D].武汉科技大学硕士论文.2009：27～31
[89]YP. Wang, L. Lees. Scale-space derived from B-splines [J]. IEEE Trans. PAMI,1998,20(10): 1040～1055
[90]梁艳.基于B样条小波变换的图像边缘检测技术研究[D].武汉理工大学硕士论文.2008：37～50
[91]Q. Jiang. Orthogonal Multiwavelets with Optimum Time-frequency Resolution [J]. IEEE Trans. SP. Special Issue on Theory and Application of Filterbanks and Wavelet Transforms,1998,46(4): 830～844
[92]S. Mallat. Multiresolution Approximation and Wavelet Orthonormal Bases of L2(R) [J]. Trans. Amer. Math. Soc,1959,315(8):59～67
[93]S. Mattlat. A theory for Multiresolution Sidecomosition:the Wavelet Representation [J]. IEEE Trans. PAMI,1989,11(7):674～693
[94]S. Mallat. Multifrequeney Channel Decomposition of Images and Wavelets [J]. IEEE Trans. ASSP,1989,37(12):2091～2110
[95]刘建军.基于小波包的边缘检测算法研究[D].中南大学硕士论文.2010：44～52