地震图像去噪算法的研究与实现
|
摘要
实际获得的图像一般都因受到某种干扰而含有噪声。引起噪声的原因有很多,如敏感元件的内部噪声、感光材料的颗粒噪声、热噪声、电器机械运动产生的抖动噪声、传输信道的干扰噪声、量化噪声等。这些噪声恶化了图像质量,使图像模糊,甚至淹没特征,给分析带来困难。
图像去噪平滑的目的就是为了减少和消除图像中的噪声,以改善图像质量,有利于抽取对图像的特征进行分析。长期以来,人们已付出许多努力,设法利用边缘信息来寻找区域,进而实现图像的识别和解释分析。在不同的应用中,人们要实现的目标不同,因此对边缘的要求也不同,目标边缘、图像纹理甚至噪声都可能成为有意义的边缘。虽然目前已研究出不少图像去噪平滑的算法,但还没有一种普遍适用于各种图像的有效方法。因此,根据具体应用要求,设计新的图像去噪平滑方法或对现有的方法进行改进,以得到满意的边缘检测结果依然是目前图像处理中研究的热点之一。
本文首先研究了经典的图像去噪平滑方法,线性及非线性滤波技术及其在图像去噪平滑中的应用,提出一种保持图像细节的自适应的去噪平滑算法,并通过理论分析以及仿真计算,比较了各种算法在图像去噪平滑中的优缺点。其次,对图像小波去噪相关算法进行研究,改进了一种中值滤波与小波去噪相结合的方法。最后,在此基础上,将改进和提出的各种算法应用到项目中,设计实现了相干切片断层多边形检测系统。通过实践验证:该系统检测出的断层多边形定位准确、连续、光滑、边缘细节丰富、边界呈单像素宽。该系统在油田地震勘探三维解释应用中取得良好检测效果,并提高了断层检测的精度和效率。
The actual image was generally due to interference which contain some noise. Noise caused many reasons, such as sensitive components of internal noise, photosensitive materials particles noise, thermal noise, electrical mechanical movement generated jitter noise, transmission channel interference noise, quantization noise. These noise worsened the image quality, image blurring, and even submerged characteristics to the analysis difficult.
Image denoising-smoothing objective is to reduce and eliminate the noise in the image, in order to improve image quality, to the characteristics of the images collected for analysis. For a long time, people have been paying many efforts to try to use the information to find the edge region, leading to the identification and interpretation of image analysis. In different applications, the people to achieve different objectives, and therefore also on the edge of the requirements of different targets marginalized, or even image texture noise may become a meaningful edge. Although many have developed image denoising-smoothing algorithm, but it does not have a generally applicable to all kinds of images effective way. Therefore, in accordance with the specific application requirements, design a new image denoising smoothing method or improve existing methods, in order to obtain satisfactory results still on the verge of image processing is a focus in the study.
This paper firstly researched into the classic image denoising-smoothing method, and linear and non-linear filtering technology and its application in image denoising-smoothing, presented a self-adaptive denoising-smoothing algorithm which can maintain image details, and compared each algorithm’s advantages and disadvantages in image denoising-smoothing through theoretical analysis and simulation. Secondly, this paper improved the method of combining median filtering with wavelet denoising based on the research of the image wavelet denoising algorithm. Finally, on the basis of the above, the various algorithms, which were improved and presented, were applied into the project, and a coherent slice fault polygon detection system was designed and implemented. The result that the fault polygon detected out by the system was position accurate, continuous, smooth, rich details of the edge, and the border showing a single pixel width was verified through practice. The system achieved good test results in the explaining and application of 3D seismic in oilfield, and improved the precision and efficiency of fault detection.
图像去噪平滑的目的就是为了减少和消除图像中的噪声,以改善图像质量,有利于抽取对图像的特征进行分析。长期以来,人们已付出许多努力,设法利用边缘信息来寻找区域,进而实现图像的识别和解释分析。在不同的应用中,人们要实现的目标不同,因此对边缘的要求也不同,目标边缘、图像纹理甚至噪声都可能成为有意义的边缘。虽然目前已研究出不少图像去噪平滑的算法,但还没有一种普遍适用于各种图像的有效方法。因此,根据具体应用要求,设计新的图像去噪平滑方法或对现有的方法进行改进,以得到满意的边缘检测结果依然是目前图像处理中研究的热点之一。
本文首先研究了经典的图像去噪平滑方法,线性及非线性滤波技术及其在图像去噪平滑中的应用,提出一种保持图像细节的自适应的去噪平滑算法,并通过理论分析以及仿真计算,比较了各种算法在图像去噪平滑中的优缺点。其次,对图像小波去噪相关算法进行研究,改进了一种中值滤波与小波去噪相结合的方法。最后,在此基础上,将改进和提出的各种算法应用到项目中,设计实现了相干切片断层多边形检测系统。通过实践验证:该系统检测出的断层多边形定位准确、连续、光滑、边缘细节丰富、边界呈单像素宽。该系统在油田地震勘探三维解释应用中取得良好检测效果,并提高了断层检测的精度和效率。
The actual image was generally due to interference which contain some noise. Noise caused many reasons, such as sensitive components of internal noise, photosensitive materials particles noise, thermal noise, electrical mechanical movement generated jitter noise, transmission channel interference noise, quantization noise. These noise worsened the image quality, image blurring, and even submerged characteristics to the analysis difficult.
Image denoising-smoothing objective is to reduce and eliminate the noise in the image, in order to improve image quality, to the characteristics of the images collected for analysis. For a long time, people have been paying many efforts to try to use the information to find the edge region, leading to the identification and interpretation of image analysis. In different applications, the people to achieve different objectives, and therefore also on the edge of the requirements of different targets marginalized, or even image texture noise may become a meaningful edge. Although many have developed image denoising-smoothing algorithm, but it does not have a generally applicable to all kinds of images effective way. Therefore, in accordance with the specific application requirements, design a new image denoising smoothing method or improve existing methods, in order to obtain satisfactory results still on the verge of image processing is a focus in the study.
This paper firstly researched into the classic image denoising-smoothing method, and linear and non-linear filtering technology and its application in image denoising-smoothing, presented a self-adaptive denoising-smoothing algorithm which can maintain image details, and compared each algorithm’s advantages and disadvantages in image denoising-smoothing through theoretical analysis and simulation. Secondly, this paper improved the method of combining median filtering with wavelet denoising based on the research of the image wavelet denoising algorithm. Finally, on the basis of the above, the various algorithms, which were improved and presented, were applied into the project, and a coherent slice fault polygon detection system was designed and implemented. The result that the fault polygon detected out by the system was position accurate, continuous, smooth, rich details of the edge, and the border showing a single pixel width was verified through practice. The system achieved good test results in the explaining and application of 3D seismic in oilfield, and improved the precision and efficiency of fault detection.
引文
[1] A.Buades, B.Coll and J.M.Morel, A non-local algorithm for image denoising, IEEE Interna-tional Conference on Computer Vision and Pattern Recognition, 2005.
[2]A.Buades,B.Coll and J.M.Morel, On Image Denoising Method, http://www.cmla.ens-cachan.frlCmla/,2004,Technical Report 2004-15, CMLA, 2004.
[3] S.Kindermann, S.Osher, and P W.Jones. Deblurring and denoising of images by nonlocal func-tionals. Multiscale Modeling&Simulation (SIAM interdisciplinary journal), 4(4):1091~1115,2005.
[4] M.Mahmoudi and G.sapiro. Fast Image and Video Denoising via Nonlocal Means of similar Neighbourhoods. IEEE signal processing letters,12:839~841, 2005.
[5] Loupas T. An Adaptive Weighted Median Filter for Speckle Suppression in Medical Ultrasonic Image[J]. IEEE Trans on Circuits System, 1989, CAS-36(1):129~135.
[6]朱菊华,杨新等.一种改进的自适应保细节中值滤波算法[J].计算机工程应用, 1001(3):93~95.
[7]梁雯,刘松林.图像加权中值滤波及分级实现方法[J].安徽大学学报(自然科学版),1997(2):42~47.
[8] Ko S. J, Lee Y. H. Center Weighted Median Filters and their Application to Image Enhancement[J]. IEEE Trans circ Syst,1991, 38(9):984~993.
[9]刘真,李绿江.用数学形态法去除图像噪声的新算法[J].北京广播学院学报(自然科学)1996(2):28~37.
[10] Heijmans T,MhenkJ.A.MorphologicalImageOperator[M].Academic print Inc,1994
[11] Witkin A. Scalespace filtering[C]. International Joint Conference on Artificial Intelligence, Karlsruhe, West Germany,1983,1091~1021.
[12] Perona P, Malik J. Scale space and edge detection using anisotropic diffusion[J]. IEEE Trans on Pattern Anal Machine Intell, 1990,12(7):629~639.
[13] Weickert J. Theoretical foundations of anisotropic diffusion in image processing[J]. Computing Suppl. 1996, 11:221~236.
[14] Weickert J. Anisotropic Diffusion in Image Processing[C]. European Consortium for Mathematics in Industry,B.G Teubner Stuttgart, 1998.
[15] Weickert J. Coherence-enhancing diffusion filtering[J]. International Journal of Computer Vision, 1999, 31(2-3):111~127.
[16] Mallat S G. A theory for multiresolution signal decomposition[J]. IEEE Trans on PAMI,1989,11(7):674~693.
[17] Mallat S, Hwang W L. Sigularity detection and processing with wavelets[J]. IEEE Trans on Information Theory,1992,38(2):617~643.
[18]何东健,耿楠,张义宽.数字图像处理[M].西安电子科技大学出版社,2003,98~99.
[19]朱秀昌,刘峰,胡栋.数字图像处理与图像通信[M].北京邮电大学出版社,2002.
[20]姚敏等编著.数字图像处理[M].北京:机械工业出版社,2006.1:106～111.
[21]陈传波,金先级等编著.数字图像处理[M].北京:机械工业出版社,2004.7:107～111.
[22]景晓军,周贤伟,付娅丽等编著.图像处理技术及其应用[M].北京:国防工业出版社,2005.8:120～150.
[23]刘禾编著.数字图像处理及应用[M].北京:中国电力出版社,2005.12:87~112.
[24]王浩,曾欢.几种图像去噪平滑算法的性能分析[J] .高新技术2006 No.3:1-2.
[25]秦志远,吴冰等.图像平滑算法比较研究及改进策略[J] .测绘学院学报2005年6月第22卷第2期.
[26] Tom ita F, T suji S. Extraction of multiple regions by smoothing in selected neighborhoods[J].IEEE Trans on Systems,Man and Cybernetics,1977, SMC-7:107~109.
[27] Richard N Czerwinski, Douglas L Jones, William D O'Brien Jr. Ultrasound speckle reduction by directional median filtering[A]. In: Proceedings of International Conference on Image Processing Vol 1 [C], Washington, D.C., 1995:358~361.
[28]刘兴洪,汪林林.一种图像边缘保持的改进方向平滑算法[J]计算机科学2006 Vol.33 No.10:210~212.
[29]李弼程,彭天强,彭波.智能图像处理技术[M] .电子工业出版社2004.7:73~75
[30] Calderbank A. R. , Daubechies I. ,Sweldens W. and Yeo B. L. Lossless image compression sing integer to integer wavelet transform In Proc. of IEEE International Conference on Image Processing , 1997.
[31]刘贵忠,张志明,冯牧等.信号重构的小波极大模整形迭代算法[M] .自然科学进展,2000.7,10(7): 660~664.
[32]李建平.小波分析与信号处理理论[M].重庆:重庆出版社,1997. 12.
[33]杨福生.小波变换的工程分析与应用[M].北京:科学出版社.1999.
[34]陈善学,郑建宏.由小波变换模极大值实现信号重构[J].重庆邮电学院学报(自然科学版),2000, (2) .
[35] Sheng F., Bilgin A. and et al Lossy and lossless image compression using reversible integer wavelet transforms. 0-8186-8821~1 /98 , IEEE.
[36] Cristea P.,Comelis J. and Munteanu A. Progessive lossless coding of medical images. Future Generation Computeu System 1998. 14: 23~32.
[37]林椹尠，宋国乡，薛文著．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(自然科学版) 2004年8月第31卷第4期627~628.
[38] Coifman R R, Donoho D L. Translatiorr invariant De-noising, in Wavelets and Statistics, Springer Lecture Notes in Statistics 103[M]. New York: Springer-Verlag, 1994. 125~150.
[39] Mallat S, A theory for multiresolution signal decomposition: the wavelet representation, [J]. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1989,11(7): 674~693.
[40] D. L.Donoho and I. M. Johnstone, Wavelet shrinkage: Asymptopia, J. R. Stat.Soc. B, 1995, 57: 301~369.
[41] Xu Yansun, Weaver J B, Healy M J et al. Wavelet Transform Domain Filters:A Spatially Selective Noise Filtration Technique [J]. IEEE Trans. Image Processing, 1994, 3(6): 743~758.
[42]李建平.小波理论与信号处理[M].重庆:重庆出版社.2001.
[43]徐朝伦.基于子波变换和模糊数学的图像分割的研究[D],北京理工大学.1998
[44] SGrace Chang Bin Yu, M Vattereli Adaptive Wavelet Threshokling for Image Denoising and Compression[J]. IEEE Trans Image Processing 2000, 9(9): 1532~1546.
[45] VattereliM, Kovacevic J Wavelet and Subband Coding[M].Englewood Cliffs N.I PrenticeHall 1995.
[46] Yang Dali Xu Mingxing Wu Wenhu A Noise Cancellation Method Based on Wavelet Transfom[C].Beijing International Symposium on Chinese Spoken Language Processing 2000. 211~214.
[47]张向君,李幼铭.三维相干切片断层多边形检测[J].物探化探计算技术.2001年11月第4期第23卷:295～298.
[2]A.Buades,B.Coll and J.M.Morel, On Image Denoising Method, http://www.cmla.ens-cachan.frlCmla/,2004,Technical Report 2004-15, CMLA, 2004.
[3] S.Kindermann, S.Osher, and P W.Jones. Deblurring and denoising of images by nonlocal func-tionals. Multiscale Modeling&Simulation (SIAM interdisciplinary journal), 4(4):1091~1115,2005.
[4] M.Mahmoudi and G.sapiro. Fast Image and Video Denoising via Nonlocal Means of similar Neighbourhoods. IEEE signal processing letters,12:839~841, 2005.
[5] Loupas T. An Adaptive Weighted Median Filter for Speckle Suppression in Medical Ultrasonic Image[J]. IEEE Trans on Circuits System, 1989, CAS-36(1):129~135.
[6]朱菊华,杨新等.一种改进的自适应保细节中值滤波算法[J].计算机工程应用, 1001(3):93~95.
[7]梁雯,刘松林.图像加权中值滤波及分级实现方法[J].安徽大学学报(自然科学版),1997(2):42~47.
[8] Ko S. J, Lee Y. H. Center Weighted Median Filters and their Application to Image Enhancement[J]. IEEE Trans circ Syst,1991, 38(9):984~993.
[9]刘真,李绿江.用数学形态法去除图像噪声的新算法[J].北京广播学院学报(自然科学)1996(2):28~37.
[10] Heijmans T,MhenkJ.A.MorphologicalImageOperator[M].Academic print Inc,1994
[11] Witkin A. Scalespace filtering[C]. International Joint Conference on Artificial Intelligence, Karlsruhe, West Germany,1983,1091~1021.
[12] Perona P, Malik J. Scale space and edge detection using anisotropic diffusion[J]. IEEE Trans on Pattern Anal Machine Intell, 1990,12(7):629~639.
[13] Weickert J. Theoretical foundations of anisotropic diffusion in image processing[J]. Computing Suppl. 1996, 11:221~236.
[14] Weickert J. Anisotropic Diffusion in Image Processing[C]. European Consortium for Mathematics in Industry,B.G Teubner Stuttgart, 1998.
[15] Weickert J. Coherence-enhancing diffusion filtering[J]. International Journal of Computer Vision, 1999, 31(2-3):111~127.
[16] Mallat S G. A theory for multiresolution signal decomposition[J]. IEEE Trans on PAMI,1989,11(7):674~693.
[17] Mallat S, Hwang W L. Sigularity detection and processing with wavelets[J]. IEEE Trans on Information Theory,1992,38(2):617~643.
[18]何东健,耿楠,张义宽.数字图像处理[M].西安电子科技大学出版社,2003,98~99.
[19]朱秀昌,刘峰,胡栋.数字图像处理与图像通信[M].北京邮电大学出版社,2002.
[20]姚敏等编著.数字图像处理[M].北京:机械工业出版社,2006.1:106～111.
[21]陈传波,金先级等编著.数字图像处理[M].北京:机械工业出版社,2004.7:107～111.
[22]景晓军,周贤伟,付娅丽等编著.图像处理技术及其应用[M].北京:国防工业出版社,2005.8:120～150.
[23]刘禾编著.数字图像处理及应用[M].北京:中国电力出版社,2005.12:87~112.
[24]王浩,曾欢.几种图像去噪平滑算法的性能分析[J] .高新技术2006 No.3:1-2.
[25]秦志远,吴冰等.图像平滑算法比较研究及改进策略[J] .测绘学院学报2005年6月第22卷第2期.
[26] Tom ita F, T suji S. Extraction of multiple regions by smoothing in selected neighborhoods[J].IEEE Trans on Systems,Man and Cybernetics,1977, SMC-7:107~109.
[27] Richard N Czerwinski, Douglas L Jones, William D O'Brien Jr. Ultrasound speckle reduction by directional median filtering[A]. In: Proceedings of International Conference on Image Processing Vol 1 [C], Washington, D.C., 1995:358~361.
[28]刘兴洪,汪林林.一种图像边缘保持的改进方向平滑算法[J]计算机科学2006 Vol.33 No.10:210~212.
[29]李弼程,彭天强,彭波.智能图像处理技术[M] .电子工业出版社2004.7:73~75
[30] Calderbank A. R. , Daubechies I. ,Sweldens W. and Yeo B. L. Lossless image compression sing integer to integer wavelet transform In Proc. of IEEE International Conference on Image Processing , 1997.
[31]刘贵忠,张志明,冯牧等.信号重构的小波极大模整形迭代算法[M] .自然科学进展,2000.7,10(7): 660~664.
[32]李建平.小波分析与信号处理理论[M].重庆:重庆出版社,1997. 12.
[33]杨福生.小波变换的工程分析与应用[M].北京:科学出版社.1999.
[34]陈善学,郑建宏.由小波变换模极大值实现信号重构[J].重庆邮电学院学报(自然科学版),2000, (2) .
[35] Sheng F., Bilgin A. and et al Lossy and lossless image compression using reversible integer wavelet transforms. 0-8186-8821~1 /98 , IEEE.
[36] Cristea P.,Comelis J. and Munteanu A. Progessive lossless coding of medical images. Future Generation Computeu System 1998. 14: 23~32.
[37]林椹尠，宋国乡，薛文著．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(自然科学版) 2004年8月第31卷第4期627~628.
[38] Coifman R R, Donoho D L. Translatiorr invariant De-noising, in Wavelets and Statistics, Springer Lecture Notes in Statistics 103[M]. New York: Springer-Verlag, 1994. 125~150.
[39] Mallat S, A theory for multiresolution signal decomposition: the wavelet representation, [J]. IEEE Trans. On Pattern Analysis and Machine Intelligence, 1989,11(7): 674~693.
[40] D. L.Donoho and I. M. Johnstone, Wavelet shrinkage: Asymptopia, J. R. Stat.Soc. B, 1995, 57: 301~369.
[41] Xu Yansun, Weaver J B, Healy M J et al. Wavelet Transform Domain Filters:A Spatially Selective Noise Filtration Technique [J]. IEEE Trans. Image Processing, 1994, 3(6): 743~758.
[42]李建平.小波理论与信号处理[M].重庆:重庆出版社.2001.
[43]徐朝伦.基于子波变换和模糊数学的图像分割的研究[D],北京理工大学.1998
[44] SGrace Chang Bin Yu, M Vattereli Adaptive Wavelet Threshokling for Image Denoising and Compression[J]. IEEE Trans Image Processing 2000, 9(9): 1532~1546.
[45] VattereliM, Kovacevic J Wavelet and Subband Coding[M].Englewood Cliffs N.I PrenticeHall 1995.
[46] Yang Dali Xu Mingxing Wu Wenhu A Noise Cancellation Method Based on Wavelet Transfom[C].Beijing International Symposium on Chinese Spoken Language Processing 2000. 211~214.
[47]张向君,李幼铭.三维相干切片断层多边形检测[J].物探化探计算技术.2001年11月第4期第23卷:295～298.