基于轮廓波变换的金属断口图像处理方法研究
|
摘要
本论文是在国家自然科学基金(51075372)和无损检测技术教育部重点实验室开放基金(No.ZD200829003)资助下,将轮廓波(Contourlet)变换应用于金属断口图像处理中,并在此基础上,深入研究了基于Contourlet变换的金属断口图像处理方法,取得了一些创新性成果。本论文的主要内容包括以下几个方面：
第一章,论述了本课题的提出及其研究意义,综述了金属断口图像的国内外研究现状,并对Contourlet变换的国内外研究现状进行了阐述,提出了本论文的主要内容及创新之处。
第二章,针对小波变换在图像处理领域中的不足,论述了Contourlet变换的基本理论和算法,同时,通过仿真实验对比了Contourlet变换和小波变换在图像处理中的优缺点,给出了离散Contourlet变换的特性。本章的内容是整篇论文的理论基础。
第三章,针对Contourlet变换在图像处理中存在的低频变换产生冗余性的不足,即Contourlet变换中的拉普拉斯分解是有冗余的,并考虑到Grouplet变换是基于图像几何流最佳稀疏表示的正交变换,可以最大限度的利用图像的几何特征,提出了一种无冗余的基于Grouplet-Contourlet变换的金属断口图像去噪和增强方法,提出的方法利用Grouplet变换代替Contourlet变换中的拉普拉斯塔形分解,消除了Contourlet变换的冗余性。同时,将提出的算法与基于传统的小波变换和Contourlet变换的图像去噪和增强算法进行了对比分析。实验结果表明,该算法优于基于传统的小波变换和Contourlet变换的图像去噪和增强算法,提高了去噪和增强图像的峰值信噪比,同时也很好的保留了图像的轮廓信息。
第四章,基于小波的Contourlet变换是一种新的无冗余和完美重构的变换,与小波变换和Contourlet变换相比较,能更好地挖掘图像的方向、纹理等信息。脉冲耦合神经网络(PCNN)是一种可视化猫激励的神经网络,具有同步脉冲发放和全局耦合等特性,它不需要学习或者训练,能从复杂背景下提取有效信息。结合两种方法的各自的优点,提出了一种基于WBCT-PCNN的图像融合方法。在提出的方法中,首先对待融合的两幅图像进行WBCT变换,然后对得到的各对应的低频和高频子带系数采用PCNN融合规则选取融合系数,最后对融合后的子带系数采用逆WBCT变换生成融合图像。同时,将提出的算法与基于PCNN.小波-PCNN、Contourlet-PCNN、NSCT-PCNN的融合算法进行了对比分析。实验结果表明,该算法在图像融合方面具有一定的优势。
第五章,将峭度的概念和Contourlet变换引入到金属断口图像处理中,并结合Contourlet变换和峭度的各自优点,给出了Contourlet峭度的定义和算法,与L1范数、平均能量相比较,Contourlet峭度能够更敏感地反映金属断口的纹理特征,并且对方向不敏感。在此基础上,提出了一种基于Contourlet峭度的金属断口图像识别方法。该方法首先对金属断口图像进行三级Contourlet变换,将各个频带输出的峭度作为断口识别的特征,输入到分类器中进行识别。同时,提出的方法还与基于L1范数和平均能量的识别方法进行了比较。试验结果表明,提出的方法是有效的。
关联向量机(relevence vector machine, RVM)具有很好的泛化能力,能对类别的归属给出一种概率度量。本章还结合RVM和Contourlet变换各自的优点,提出了一种基于Contourlet-RVM金属断口图像识别方法。在提出的方法中,首先对金属断口图像进行两级Contourlet变换,将各个频带输出的L1范数作为断口识别的特征,然后采用RVM分类器进行分类。同时,提出的方法还与Contourlet-SVM识别方法进行了对比分析。实验结果表明,提出的方法是有效的,不论在正确识别率方面,还是在训练速度方面,Contourlet-RVM识别方法都优于Contourlet-SVM识别方法。
第六章,对本论文的研究工作进行了总结,并提出了值得进一步研究的问题。
This thesis is supported by the National Natural Science Foundation of China (No.51075372) and the open Fund of Key Laboratory of the Ministry of Education of Nondestructive Testing technology (No.ZD200829003), introduces Contourlet transform to the metal fracture image processing, such as image denoising, enhancement, fusion and recognition. Some preferable innovative achievements are obtained in this paper. The primary contents of this article involve the followings:
Chapter one illuminates the significance of proposing and studying on this thesis, summarizes the research status at home and abroad of the metal fracture image processing and Contourlet transform, and presentes the primary contents and innovation points of this paper at last.
Chapter two introduces the theory of the Contourlet transform due to the shortages of wavelet transform in image processing area. At the same time, the advantages and disadvantages of the Contourlet transform and the wavelet tranform are compared, and the characteristics of the pyramidal directional filter bank Contourlet are given. The content of this chapter is the basic theory of the whole thesis.
Chapter three introduces an algorithm based on non-redundancy Grouplet-Contourlet transform in image denoising and enhancement as the low-frequency decomposition of the Contourlet transform has a redundancy up to33%and the new transform Grouplet proposed by Mallat in2008, which is implemented with an orthogonal weighted Haar transform that adapts the lifting parameters to groupings specified by multiscale association fields, can exploit the geometry of the image. At the same time, the proposed method is compared with the methods based on wavelet transform and Contourlet transform in image denoising and enhancement. The experimental results show that the proposed method is superior to other methods. It not only improves the PSNR of the denoised and enhanced image, but also retains good contour information of the image.
Chapter four combining a non-redundancy algorithm based on WBCT—a new non-redundant and perfect reconstruction transform, which can exploit more information of the image's direction and texture and the latest research results of the Biology and Neuroscience in recent years—PCNN, which can extract useful information from complex background and does not require learning or training, proposes a fusion algorithm based on WBCT-PCNN. In the proposed method, firstly, two original images were decomposed by using WBCT. Then, the low frequency subband coefficients and a series of bandpass directional subband coefficients were obtained. All of fused subband coefficients were determined by image fusion algorithm based on PCNN. Finally, by performing the inverse WBCT on the fused subband coefficients, the fused image was obtained. The experimental results show that the fusion method is superior to fusion methods based on PCNN, wavelet-PCNN, Contourlet-PCNN and NSCT-PCNN.
Chapter five introduces kurtosis into the field of image processing and defines the concept of Contourlet kurtosis. Kurtosis is more sensitive to the texture of the metal fracture than L1norm and the average energy, while it is not sensitive to the direction of the image. On this basis, a recognition method of metal fracture image based on Contourlet kurtosis is proposed. In the proposed method, three-level Contourlet transform is applied to each metal fracture image firstly, and the kurtosis of each Contourlet transform frequency band output is used as the characteristics of fracture image recognition. Then K-nearest neighbor classifier is used to classify the fracture images. The experimental results show that the proposed method is superior to the recognition method based on L1norm and the average energy calculated by Contourlet coefficient.
The relevance vector machine has a good generalization ability and can give a probability measure on the ownership of the categories. Combining the respective advantage of Contourlet transform and RVM, a recognition method of metal fracture image of aerial material based on Contourlet-RVM is also proposed in this chapter. In the proposed method, two-lever wavelet transform is applied to metal fracture image recognition by selecting Contourlet transform. L1norm of each wavelet transform frequency band output is used as the characteristics of fracture image recognition. And RVM classifier is used to classify the fracture image. At the same time, the proposed method is compared with the Contourlet-SVM recognition method. The experiment result shows that the proposed method is very effective. Whether in the correct recognition rate, or in the training speed, the Contourlet-RVM recognition method is superior to the Contourlet-SVM recognition method.
第一章,论述了本课题的提出及其研究意义,综述了金属断口图像的国内外研究现状,并对Contourlet变换的国内外研究现状进行了阐述,提出了本论文的主要内容及创新之处。
第二章,针对小波变换在图像处理领域中的不足,论述了Contourlet变换的基本理论和算法,同时,通过仿真实验对比了Contourlet变换和小波变换在图像处理中的优缺点,给出了离散Contourlet变换的特性。本章的内容是整篇论文的理论基础。
第三章,针对Contourlet变换在图像处理中存在的低频变换产生冗余性的不足,即Contourlet变换中的拉普拉斯分解是有冗余的,并考虑到Grouplet变换是基于图像几何流最佳稀疏表示的正交变换,可以最大限度的利用图像的几何特征,提出了一种无冗余的基于Grouplet-Contourlet变换的金属断口图像去噪和增强方法,提出的方法利用Grouplet变换代替Contourlet变换中的拉普拉斯塔形分解,消除了Contourlet变换的冗余性。同时,将提出的算法与基于传统的小波变换和Contourlet变换的图像去噪和增强算法进行了对比分析。实验结果表明,该算法优于基于传统的小波变换和Contourlet变换的图像去噪和增强算法,提高了去噪和增强图像的峰值信噪比,同时也很好的保留了图像的轮廓信息。
第四章,基于小波的Contourlet变换是一种新的无冗余和完美重构的变换,与小波变换和Contourlet变换相比较,能更好地挖掘图像的方向、纹理等信息。脉冲耦合神经网络(PCNN)是一种可视化猫激励的神经网络,具有同步脉冲发放和全局耦合等特性,它不需要学习或者训练,能从复杂背景下提取有效信息。结合两种方法的各自的优点,提出了一种基于WBCT-PCNN的图像融合方法。在提出的方法中,首先对待融合的两幅图像进行WBCT变换,然后对得到的各对应的低频和高频子带系数采用PCNN融合规则选取融合系数,最后对融合后的子带系数采用逆WBCT变换生成融合图像。同时,将提出的算法与基于PCNN.小波-PCNN、Contourlet-PCNN、NSCT-PCNN的融合算法进行了对比分析。实验结果表明,该算法在图像融合方面具有一定的优势。
第五章,将峭度的概念和Contourlet变换引入到金属断口图像处理中,并结合Contourlet变换和峭度的各自优点,给出了Contourlet峭度的定义和算法,与L1范数、平均能量相比较,Contourlet峭度能够更敏感地反映金属断口的纹理特征,并且对方向不敏感。在此基础上,提出了一种基于Contourlet峭度的金属断口图像识别方法。该方法首先对金属断口图像进行三级Contourlet变换,将各个频带输出的峭度作为断口识别的特征,输入到分类器中进行识别。同时,提出的方法还与基于L1范数和平均能量的识别方法进行了比较。试验结果表明,提出的方法是有效的。
关联向量机(relevence vector machine, RVM)具有很好的泛化能力,能对类别的归属给出一种概率度量。本章还结合RVM和Contourlet变换各自的优点,提出了一种基于Contourlet-RVM金属断口图像识别方法。在提出的方法中,首先对金属断口图像进行两级Contourlet变换,将各个频带输出的L1范数作为断口识别的特征,然后采用RVM分类器进行分类。同时,提出的方法还与Contourlet-SVM识别方法进行了对比分析。实验结果表明,提出的方法是有效的,不论在正确识别率方面,还是在训练速度方面,Contourlet-RVM识别方法都优于Contourlet-SVM识别方法。
第六章,对本论文的研究工作进行了总结,并提出了值得进一步研究的问题。
This thesis is supported by the National Natural Science Foundation of China (No.51075372) and the open Fund of Key Laboratory of the Ministry of Education of Nondestructive Testing technology (No.ZD200829003), introduces Contourlet transform to the metal fracture image processing, such as image denoising, enhancement, fusion and recognition. Some preferable innovative achievements are obtained in this paper. The primary contents of this article involve the followings:
Chapter one illuminates the significance of proposing and studying on this thesis, summarizes the research status at home and abroad of the metal fracture image processing and Contourlet transform, and presentes the primary contents and innovation points of this paper at last.
Chapter two introduces the theory of the Contourlet transform due to the shortages of wavelet transform in image processing area. At the same time, the advantages and disadvantages of the Contourlet transform and the wavelet tranform are compared, and the characteristics of the pyramidal directional filter bank Contourlet are given. The content of this chapter is the basic theory of the whole thesis.
Chapter three introduces an algorithm based on non-redundancy Grouplet-Contourlet transform in image denoising and enhancement as the low-frequency decomposition of the Contourlet transform has a redundancy up to33%and the new transform Grouplet proposed by Mallat in2008, which is implemented with an orthogonal weighted Haar transform that adapts the lifting parameters to groupings specified by multiscale association fields, can exploit the geometry of the image. At the same time, the proposed method is compared with the methods based on wavelet transform and Contourlet transform in image denoising and enhancement. The experimental results show that the proposed method is superior to other methods. It not only improves the PSNR of the denoised and enhanced image, but also retains good contour information of the image.
Chapter four combining a non-redundancy algorithm based on WBCT—a new non-redundant and perfect reconstruction transform, which can exploit more information of the image's direction and texture and the latest research results of the Biology and Neuroscience in recent years—PCNN, which can extract useful information from complex background and does not require learning or training, proposes a fusion algorithm based on WBCT-PCNN. In the proposed method, firstly, two original images were decomposed by using WBCT. Then, the low frequency subband coefficients and a series of bandpass directional subband coefficients were obtained. All of fused subband coefficients were determined by image fusion algorithm based on PCNN. Finally, by performing the inverse WBCT on the fused subband coefficients, the fused image was obtained. The experimental results show that the fusion method is superior to fusion methods based on PCNN, wavelet-PCNN, Contourlet-PCNN and NSCT-PCNN.
Chapter five introduces kurtosis into the field of image processing and defines the concept of Contourlet kurtosis. Kurtosis is more sensitive to the texture of the metal fracture than L1norm and the average energy, while it is not sensitive to the direction of the image. On this basis, a recognition method of metal fracture image based on Contourlet kurtosis is proposed. In the proposed method, three-level Contourlet transform is applied to each metal fracture image firstly, and the kurtosis of each Contourlet transform frequency band output is used as the characteristics of fracture image recognition. Then K-nearest neighbor classifier is used to classify the fracture images. The experimental results show that the proposed method is superior to the recognition method based on L1norm and the average energy calculated by Contourlet coefficient.
The relevance vector machine has a good generalization ability and can give a probability measure on the ownership of the categories. Combining the respective advantage of Contourlet transform and RVM, a recognition method of metal fracture image of aerial material based on Contourlet-RVM is also proposed in this chapter. In the proposed method, two-lever wavelet transform is applied to metal fracture image recognition by selecting Contourlet transform. L1norm of each wavelet transform frequency band output is used as the characteristics of fracture image recognition. And RVM classifier is used to classify the fracture image. At the same time, the proposed method is compared with the Contourlet-SVM recognition method. The experiment result shows that the proposed method is very effective. Whether in the correct recognition rate, or in the training speed, the Contourlet-RVM recognition method is superior to the Contourlet-SVM recognition method.
引文
[1]Zapffe C A, Clogg M. Fractography—a new tool for metallurgical research[J]. Trans. Am. Soc. Metals,1945,34:71-101.
[2]谢燕江.中值滤波在图像去噪中的应用[J].湘南学院学报,2004,25(02)：75-77.
[3]胥妍.基于小波变换技术的图像融合方法的研究与应用[D].山东师范大学,2008.
[4]刘晓召.基于小波变换的纹理图像多尺度分割算法研究[D].重庆大学,2010.
[5]周旋,周树道,黄峰,周小滔.基于小波变换的图像增强新算法[J]计算机应用,2005,25(03)：606-608.
[6]MANDELBROT B B, PASSOLA D E, PALLAYA J. Fractal character of fracture surfaces of meatals [J].Nature.1984,308(19):721-726.
[7]K Minoshima, T Nagasaki, K Komai. Automatic Classification of Fracture Surface Morphology Using Computer Image Processing Technique, Trans.Jpn.Soc.Mech.Eng, 1990,56(525):1318-1328.
[8]K. Komai, K. Minoshima and S. Ishii, Recognition of different fracture surface morphologies using computer image processing technique. JSME Int. J., Series A,1993,36 (2):220-227.
[9]马朋仁.基于混合向量和支持向量机的金属断口图像识别与分类[东北大学硕士学位文].2004.
[10]颜云辉,马朋仁,曹宇光,张坚.金属断口图像句法模式识别与分类方法[J].中国机械工程,2004,15(3)：259-263.
[11]颜云辉,王德俊,黄雨华,王清吴,王彤波.疲劳断口分析的Fourier变换方法.金属学报,1997,33(4)：386-390.
[12]张晓庆,孙永庆,张峥.基于不确定性纹理谱的金属断口形貌识别.金属学报,2004,40(10)：1018-1022.
[13]苏静,黎明.基于灰度共生矩阵的金属断口图像的分类研究.计算机工程与应用,2008,44(9)：223-225.
[14]李沁,田沛.应用分行进行金属断口的识别分类,仪器仪表与分析检测.2005,4：19-21.
[15]颜云辉,马朋仁,曹宇光,张坚.金属断口图像句法模式识别与分类方法.中国机械工程,2004,15(3)：259-263.线性最小距离分类
[16]颜云辉,杨会林,王成明.金属断口图像的非线性模式识别方法.东北火学学报(自然科学版),2004,25(9)：884-886.
[17]张莉,黎明,杨小芹.基于树形小波变换的断口图像识别[J].南昌航空大学学报：自然科学版,2007,21(2)：42-45.
[18]颜云辉,杨会林,王成明,雷世超.基于神经网络的金属断口图像识别信息知识库的构建方法. http://www.paper.edu.cn/downloadpaper.php?serial_number=200411-33&type=1
[19]M.N, Do and M. Vetterli, Contourlets:a New Directional Multiresolution Image Representation.Signal,Systems and Computers,2002,1:497-501.
[20]王晅,张小景,马进明Contourlet域中邻域窗最优闽值滤噪算法.计算机工程,2010,36(5)：223-227.
[21]李绿森.基于Contourlet变换的图像增强算法[硕士学位论文].厦门大学,2008.
[22]沙宇恒,丛琳,孙强,焦李成.基于Contourlet域HMT模型的多尺度图像分割.红外与毫米波学报,2005,12：472-476.
[23]肖羽,王相海.基于Contourlet的中低码率图像质量可分级编码算法.计算机研究与发展,2008,45(6)：1020-1028.
[24]薛旭成,张淑艳,李洪法,郭永飞Contourlet变换在图像压缩中的应用研究.微计算机信息,2009,25(3-3)：266-270.
[25]屈小波,闫敬文,杨贵德.改进拉普拉斯能量和的尖锐频率局部化Contourlet域多聚焦图像融合方法.光学精密工程,2009,17(5)：1023-1211.
[26]章立,彭宏京.基于Contourlet变换的纹理特征提取及纹理分类.计算机工程与设计,2010,31(10)：2343-2346.
[27]石一兴,吴炜,罗代升,喻继业.基于Contourlet域的遥感图像超分辨率复原技术.计算机应用,2010,30(4)：939-942.
[28]刘灿,刘哲,徐华楠.循环平移和非下采样在轮廓波去噪中的应用.计算机工程与应用,2009,45(33)：144-152.
[29]冯鹏,魏彪,潘英俊,米德伶,金炜.基于方向滤波器组的Contourlet变换频谱混叠特性研究.光电子·激光,2008,19(12)：1670-1674.
[30]金炜,励金祥,杨任尔.抗混叠轮廓波HMT模型的医学图像融合.中国图象图形学报,2010,15(2)：321-327.
[31]闫河,刘加伶,闫卫军,冯鹏,张小川.基于去频谱混叠Contourlet变换的层内局部相关性图像降噪.电路与系统学报,2009,14(1)：82-86.
[32]章立,彭宏京.基于WBCT的纹理分类算法.计算机工程与设计,2010,31(23)：5056-5059.
[33]向丽.基于WBCT与平滑共生矩阵的图像检索.通信技术,2009,42(12)：150-152.
[34]宋亚军,倪国强,高昆.基于小波-Contourlet变换的区域能量加权图像融合算法.北京理工大学学报,2008,28(2)：168-172.
[35]殷兵,于梅,梁栋,鲍文霞.基于小波-Contourlet变换的区域能量融合规则的图像融合算法.计算机应用研究,2008,25(7)：2197-2199.
[36]吕嘉凯,冯燕,吕维Contourlet和小波相结合的SPIHT算法.计算机工程与应用,2009,45(27)：168-170.
[37]孙文方,宋蓓蓓,赵亦工联合Contourlet和Bandelet的变换在图像压缩中的应用.西安电子科技大学学报.2007,34(4)：611-615.
[38]郑宇杰,胡文宁,吴小俊.基于Contourlet域主成分分析的SAR图像去噪.计算机仿真.2009,26(6)：242-245.
[39]焦李成,侯彪等.图像多尺度几何分析理论与应用一后小波分析理论与应用[M].西安：西安电子科技大学出版社,2008.7：9-12
[40]M.N.Do, Directional Multiresolution Image Representations, Ph.D Desertation October 23, 2001.
[41]焦李成,谭山.图像多尺度几何分析：回顾和展望[J].电子学报,2003,31(12)：1975-1981
[42]P.J.Burt, E.H. Adelson. The Laplacian pyramid as a compact image code. Traps. Communication,1983,31(4):532-540.
[43]Bamberger R H, Smith M J T. A filter bank for the directional decomposition of images: Theory and design [J]. IEEE Transactions on Signal Processing,1992,40(4):882-893.
[44]谢美华,王正明.基于边缘定向扩散的图像增强方法[J].光子学报,2005,34(9)：1420-1424.
[45]王龙,汪天富,郑昌琼,李德玉,尹光福.基于灰度直方图均衡的超声医学图像增强方法[J]四川大学学报(工程科学版),2002,34(01)：105-108.
[46]王红霞,成礼智,吴翊.基于二进小波变换的噪声图像自适应增强新算法[J].国防科技大学学报,2005,27(1)：51-59.
[47]S. Mallat, "Geometrical grouplets," to Appear in Applied and Computational Harmonic Analysis,2008.
[48]殷兵,于梅,方洁,梁栋.基于小波-Contourlet变换的图像增强算法[J].计算机工程与设计,2008,29(12)：3142-3144.
[49]任洪娥,王海丰,赵鹏.基于小波-Contourlet变换的Cycle spinning硬阈值图像去噪方法[J]光学技术,2008,34(6)：854-857.
[50]于梅,殷兵,何国栋,梁栋.一种基于小波-Contourlet变换的图像去噪算法[J].微电子学与计算机,2008,25(12)：100-102.
[51]向静波,苏秀琴,陆陶.基于Contourlet变换和形态学的图像增强方法[J].光子学报,2009,38(01)：224-227.
[52]邱宣,周则明,胡友彬.应用邻域方差加权平均的curvelet变换融合[J].计算机工程与应用,2010,46(9)：166-168.
[53]王永森,谭景信.一种基于小波变换的多尺度图像融合方法[J].计算机工程与设计,2004,25(10)：1840-1842.
[54]刘和祥,冯新喜,王君.基于自组织神经网络的多传感器遥感图像融合技术[J].传感器技术,2004,23(12)：14-16.
[55]Eslami R, Radha H. Wavelet-based Contourlet Coding Using an SPIHT-like Algorith[C]. In: Proc. IEEE International Conference on Image Processing. Piscateway, USA,2004: 784-788.
[56]Eslami R, Radha H. Wavelet based contourlet transform and its application to image coding [C]. In:Proc. IEEE International Conference on Image Processing. Piscateway, USA,2004: 3189-3192.
[57]Eckhorn R, Reitboeck H J, Arndt M, et al. Feature linking via synchronization among distributed assemblies:Simulations of results from cat visual cortex [J]. Neural Computation(S0899-7667),1990,2(3):293-307.
[58]Cun ha A L, Zhou Jianping, Do M N. The Nonsubsampled Contourlet Transform:Theory, Design, and Applications [J]. IEEE Trans on Image Processing,2006,15(10):3089-3101.
[59]Dicke P W. Simulation dymanischer Merkmalskopplungen in einem neuronalen Netzwerkmodell[D]. Ph.D.dissertation, philipps Univ., Marburg, Germany,1992.
[60]王海晖,彭嘉雄,吴巍,李峰.多源遥感图像融合效果评价方法研究[J].计算机工程与应用,2003,25(5)：33-37.
[61]李富才,孟光,胥永刚.基于多尺度峭度指标的损伤识别研究[J].振动与冲击,2010,29(S)：76-78.
[62]袁云龙.基于峭度-小波包分析的滚动轴承故障诊断[J].新技术新工艺,2008,(05)：43-46.
[63]陈军堂,廖世勇,甘剑锋,米林,达峰.峭度分析法在发动机噪声信号故障特征提取中的应用[J].移动电源与车辆,2011,(1)：15-18.
[64]蒋瑜,陈循,陶俊勇,张春华.指定功率谱密度、偏斜度和峭度值下的非高斯随机过程数字模拟[J].系统仿真学报,2006,18(5)：1127-1130.
[65]毋文峰,陈小虎,苏勋家,王旭平,姚春江.基于峭度的ICA特征提取和齿轮泵故障诊断[J].机械科学与技术,2011,30(9)：1583-1587.
[66]陈寿齐,沈越泓,马明,许魁.基于峭度的稳健特定信号盲提取[J].系统仿真学报,2008,20(2)：6093-6096.
[67]郑红岩,李红波,曾繁景,李铁峰.基于峭度和负熵的模拟电路故障诊断方法研究[J].信息工程大学学报,2011,12(2)：203-207.
[68]Tipping M E. The relevance vector machine[C]//Proc. of Advances in Neural Information Processing Systems. [S.1.]:MIT Press,2000:652-658.
[69]Xiang Minxu, Mao Yunfeng, Xiong Jiani, et al. Classification performance comparison between RVM and SVM [C]. IEEE International Workshop on Anti-counterfeiting, Security, Identification. USA:IEEE,2007:208-211.
[70]Do M N, Vetterli M. Contourlets:the contourlet transform:an efficient directional multiresolution image representation[J]. IEEE Transactions on Image Processing,2005, 14(12):2091-2106.
[71]Kenta Yamagiwa, Satoshi Izumi, Shinsuke Sakai. Detecting method of striation region of fatigue fracture surface using wavelet transform[J]. Journal of the Society of Materials Science, Japan,2004,53(3):306-312.
[72]徐飞.金属断口SEM图像的特征提取与分析[D].成都电子科技大学,2006.
[73]范伊红,李敏,张元.相关向量机在车型识别中的应用研究.计算机工程与设计[J],2008,29(6)：110-1515.
[74]羌海益.基于Contourlet变换与SVM的人脸识别方法研究ACADEMIC RESEARCH学术研究,2009：100-103.
[2]谢燕江.中值滤波在图像去噪中的应用[J].湘南学院学报,2004,25(02)：75-77.
[3]胥妍.基于小波变换技术的图像融合方法的研究与应用[D].山东师范大学,2008.
[4]刘晓召.基于小波变换的纹理图像多尺度分割算法研究[D].重庆大学,2010.
[5]周旋,周树道,黄峰,周小滔.基于小波变换的图像增强新算法[J]计算机应用,2005,25(03)：606-608.
[6]MANDELBROT B B, PASSOLA D E, PALLAYA J. Fractal character of fracture surfaces of meatals [J].Nature.1984,308(19):721-726.
[7]K Minoshima, T Nagasaki, K Komai. Automatic Classification of Fracture Surface Morphology Using Computer Image Processing Technique, Trans.Jpn.Soc.Mech.Eng, 1990,56(525):1318-1328.
[8]K. Komai, K. Minoshima and S. Ishii, Recognition of different fracture surface morphologies using computer image processing technique. JSME Int. J., Series A,1993,36 (2):220-227.
[9]马朋仁.基于混合向量和支持向量机的金属断口图像识别与分类[东北大学硕士学位文].2004.
[10]颜云辉,马朋仁,曹宇光,张坚.金属断口图像句法模式识别与分类方法[J].中国机械工程,2004,15(3)：259-263.
[11]颜云辉,王德俊,黄雨华,王清吴,王彤波.疲劳断口分析的Fourier变换方法.金属学报,1997,33(4)：386-390.
[12]张晓庆,孙永庆,张峥.基于不确定性纹理谱的金属断口形貌识别.金属学报,2004,40(10)：1018-1022.
[13]苏静,黎明.基于灰度共生矩阵的金属断口图像的分类研究.计算机工程与应用,2008,44(9)：223-225.
[14]李沁,田沛.应用分行进行金属断口的识别分类,仪器仪表与分析检测.2005,4：19-21.
[15]颜云辉,马朋仁,曹宇光,张坚.金属断口图像句法模式识别与分类方法.中国机械工程,2004,15(3)：259-263.线性最小距离分类
[16]颜云辉,杨会林,王成明.金属断口图像的非线性模式识别方法.东北火学学报(自然科学版),2004,25(9)：884-886.
[17]张莉,黎明,杨小芹.基于树形小波变换的断口图像识别[J].南昌航空大学学报：自然科学版,2007,21(2)：42-45.
[18]颜云辉,杨会林,王成明,雷世超.基于神经网络的金属断口图像识别信息知识库的构建方法. http://www.paper.edu.cn/downloadpaper.php?serial_number=200411-33&type=1
[19]M.N, Do and M. Vetterli, Contourlets:a New Directional Multiresolution Image Representation.Signal,Systems and Computers,2002,1:497-501.
[20]王晅,张小景,马进明Contourlet域中邻域窗最优闽值滤噪算法.计算机工程,2010,36(5)：223-227.
[21]李绿森.基于Contourlet变换的图像增强算法[硕士学位论文].厦门大学,2008.
[22]沙宇恒,丛琳,孙强,焦李成.基于Contourlet域HMT模型的多尺度图像分割.红外与毫米波学报,2005,12：472-476.
[23]肖羽,王相海.基于Contourlet的中低码率图像质量可分级编码算法.计算机研究与发展,2008,45(6)：1020-1028.
[24]薛旭成,张淑艳,李洪法,郭永飞Contourlet变换在图像压缩中的应用研究.微计算机信息,2009,25(3-3)：266-270.
[25]屈小波,闫敬文,杨贵德.改进拉普拉斯能量和的尖锐频率局部化Contourlet域多聚焦图像融合方法.光学精密工程,2009,17(5)：1023-1211.
[26]章立,彭宏京.基于Contourlet变换的纹理特征提取及纹理分类.计算机工程与设计,2010,31(10)：2343-2346.
[27]石一兴,吴炜,罗代升,喻继业.基于Contourlet域的遥感图像超分辨率复原技术.计算机应用,2010,30(4)：939-942.
[28]刘灿,刘哲,徐华楠.循环平移和非下采样在轮廓波去噪中的应用.计算机工程与应用,2009,45(33)：144-152.
[29]冯鹏,魏彪,潘英俊,米德伶,金炜.基于方向滤波器组的Contourlet变换频谱混叠特性研究.光电子·激光,2008,19(12)：1670-1674.
[30]金炜,励金祥,杨任尔.抗混叠轮廓波HMT模型的医学图像融合.中国图象图形学报,2010,15(2)：321-327.
[31]闫河,刘加伶,闫卫军,冯鹏,张小川.基于去频谱混叠Contourlet变换的层内局部相关性图像降噪.电路与系统学报,2009,14(1)：82-86.
[32]章立,彭宏京.基于WBCT的纹理分类算法.计算机工程与设计,2010,31(23)：5056-5059.
[33]向丽.基于WBCT与平滑共生矩阵的图像检索.通信技术,2009,42(12)：150-152.
[34]宋亚军,倪国强,高昆.基于小波-Contourlet变换的区域能量加权图像融合算法.北京理工大学学报,2008,28(2)：168-172.
[35]殷兵,于梅,梁栋,鲍文霞.基于小波-Contourlet变换的区域能量融合规则的图像融合算法.计算机应用研究,2008,25(7)：2197-2199.
[36]吕嘉凯,冯燕,吕维Contourlet和小波相结合的SPIHT算法.计算机工程与应用,2009,45(27)：168-170.
[37]孙文方,宋蓓蓓,赵亦工联合Contourlet和Bandelet的变换在图像压缩中的应用.西安电子科技大学学报.2007,34(4)：611-615.
[38]郑宇杰,胡文宁,吴小俊.基于Contourlet域主成分分析的SAR图像去噪.计算机仿真.2009,26(6)：242-245.
[39]焦李成,侯彪等.图像多尺度几何分析理论与应用一后小波分析理论与应用[M].西安：西安电子科技大学出版社,2008.7：9-12
[40]M.N.Do, Directional Multiresolution Image Representations, Ph.D Desertation October 23, 2001.
[41]焦李成,谭山.图像多尺度几何分析：回顾和展望[J].电子学报,2003,31(12)：1975-1981
[42]P.J.Burt, E.H. Adelson. The Laplacian pyramid as a compact image code. Traps. Communication,1983,31(4):532-540.
[43]Bamberger R H, Smith M J T. A filter bank for the directional decomposition of images: Theory and design [J]. IEEE Transactions on Signal Processing,1992,40(4):882-893.
[44]谢美华,王正明.基于边缘定向扩散的图像增强方法[J].光子学报,2005,34(9)：1420-1424.
[45]王龙,汪天富,郑昌琼,李德玉,尹光福.基于灰度直方图均衡的超声医学图像增强方法[J]四川大学学报(工程科学版),2002,34(01)：105-108.
[46]王红霞,成礼智,吴翊.基于二进小波变换的噪声图像自适应增强新算法[J].国防科技大学学报,2005,27(1)：51-59.
[47]S. Mallat, "Geometrical grouplets," to Appear in Applied and Computational Harmonic Analysis,2008.
[48]殷兵,于梅,方洁,梁栋.基于小波-Contourlet变换的图像增强算法[J].计算机工程与设计,2008,29(12)：3142-3144.
[49]任洪娥,王海丰,赵鹏.基于小波-Contourlet变换的Cycle spinning硬阈值图像去噪方法[J]光学技术,2008,34(6)：854-857.
[50]于梅,殷兵,何国栋,梁栋.一种基于小波-Contourlet变换的图像去噪算法[J].微电子学与计算机,2008,25(12)：100-102.
[51]向静波,苏秀琴,陆陶.基于Contourlet变换和形态学的图像增强方法[J].光子学报,2009,38(01)：224-227.
[52]邱宣,周则明,胡友彬.应用邻域方差加权平均的curvelet变换融合[J].计算机工程与应用,2010,46(9)：166-168.
[53]王永森,谭景信.一种基于小波变换的多尺度图像融合方法[J].计算机工程与设计,2004,25(10)：1840-1842.
[54]刘和祥,冯新喜,王君.基于自组织神经网络的多传感器遥感图像融合技术[J].传感器技术,2004,23(12)：14-16.
[55]Eslami R, Radha H. Wavelet-based Contourlet Coding Using an SPIHT-like Algorith[C]. In: Proc. IEEE International Conference on Image Processing. Piscateway, USA,2004: 784-788.
[56]Eslami R, Radha H. Wavelet based contourlet transform and its application to image coding [C]. In:Proc. IEEE International Conference on Image Processing. Piscateway, USA,2004: 3189-3192.
[57]Eckhorn R, Reitboeck H J, Arndt M, et al. Feature linking via synchronization among distributed assemblies:Simulations of results from cat visual cortex [J]. Neural Computation(S0899-7667),1990,2(3):293-307.
[58]Cun ha A L, Zhou Jianping, Do M N. The Nonsubsampled Contourlet Transform:Theory, Design, and Applications [J]. IEEE Trans on Image Processing,2006,15(10):3089-3101.
[59]Dicke P W. Simulation dymanischer Merkmalskopplungen in einem neuronalen Netzwerkmodell[D]. Ph.D.dissertation, philipps Univ., Marburg, Germany,1992.
[60]王海晖,彭嘉雄,吴巍,李峰.多源遥感图像融合效果评价方法研究[J].计算机工程与应用,2003,25(5)：33-37.
[61]李富才,孟光,胥永刚.基于多尺度峭度指标的损伤识别研究[J].振动与冲击,2010,29(S)：76-78.
[62]袁云龙.基于峭度-小波包分析的滚动轴承故障诊断[J].新技术新工艺,2008,(05)：43-46.
[63]陈军堂,廖世勇,甘剑锋,米林,达峰.峭度分析法在发动机噪声信号故障特征提取中的应用[J].移动电源与车辆,2011,(1)：15-18.
[64]蒋瑜,陈循,陶俊勇,张春华.指定功率谱密度、偏斜度和峭度值下的非高斯随机过程数字模拟[J].系统仿真学报,2006,18(5)：1127-1130.
[65]毋文峰,陈小虎,苏勋家,王旭平,姚春江.基于峭度的ICA特征提取和齿轮泵故障诊断[J].机械科学与技术,2011,30(9)：1583-1587.
[66]陈寿齐,沈越泓,马明,许魁.基于峭度的稳健特定信号盲提取[J].系统仿真学报,2008,20(2)：6093-6096.
[67]郑红岩,李红波,曾繁景,李铁峰.基于峭度和负熵的模拟电路故障诊断方法研究[J].信息工程大学学报,2011,12(2)：203-207.
[68]Tipping M E. The relevance vector machine[C]//Proc. of Advances in Neural Information Processing Systems. [S.1.]:MIT Press,2000:652-658.
[69]Xiang Minxu, Mao Yunfeng, Xiong Jiani, et al. Classification performance comparison between RVM and SVM [C]. IEEE International Workshop on Anti-counterfeiting, Security, Identification. USA:IEEE,2007:208-211.
[70]Do M N, Vetterli M. Contourlets:the contourlet transform:an efficient directional multiresolution image representation[J]. IEEE Transactions on Image Processing,2005, 14(12):2091-2106.
[71]Kenta Yamagiwa, Satoshi Izumi, Shinsuke Sakai. Detecting method of striation region of fatigue fracture surface using wavelet transform[J]. Journal of the Society of Materials Science, Japan,2004,53(3):306-312.
[72]徐飞.金属断口SEM图像的特征提取与分析[D].成都电子科技大学,2006.
[73]范伊红,李敏,张元.相关向量机在车型识别中的应用研究.计算机工程与设计[J],2008,29(6)：110-1515.
[74]羌海益.基于Contourlet变换与SVM的人脸识别方法研究ACADEMIC RESEARCH学术研究,2009：100-103.