显微组织数字图像处理及自动测试
|
摘要
显微硬度计是用来分析材料表面硬度、分析金相组织的一个常用工具。各种显微硬度计品种繁多，它们在各自的实现方式以及功能方面也略有差别。控制系统的自动化程度、对金相参量进行自动分析方面是影响整个系统效率和精度的主要因素。目前，在这些方面普遍有所欠缺。
小波变换被称为是数学上的显微镜，具有很好的时、频定位能力。在正交小波变换理论中具有重要地位的多分辨率分析理论，为我们提供了一个多分辨率逼近信号和观察信号细节的工具。小波分析与现代数字图像技术的发展，为计算机分析金相参量提供了强有力的分析工具。本文的主要研究内容就是围绕提高显微硬度计测量系统的信息化、自动化程度中的几个关键问题进行的：一是载物台的数控问题；二是金相图片尺寸的标定；三是利用数字图像处理技术对金相参数进行自动测量；四是对试验压痕边缘进行快速有效地提取，解决了一般数字图像处理技术难以提取其边缘的难题，从而实现硬度的自动计算；最后是自动调焦系统。本文分别从理论和实际应用的角度对其中的难点进行了深入地分析，在几个方面取得了创新。
本文运用基于先验知识的小波多分辨率理论对金相试样上的试验压痕图像进行处理，提出了分析策略，进行了深入的理论研究和实际分析对比，实现对压痕边缘快速而有效的提取，从而实现硬度的自动分析计算，本算法是一个具有特色的创新点。
基于区域生长和数学形态学的第二相相对含量测定技术也是本文在自动分析金相参数应用中具有创新意义的技术，它的灵活性和准确率都比较高；另外，本文还提出了基于直方图技术的金相组织第二相相对含量的测定，并对这两种方法的优劣、应用场合进行了对比分析。
本文解决了金相图片尺寸的标定，提出了对金相组织单个晶粒直径、晶粒平均大小的测量方法。
本文在总结现有自动调焦技术的基础上，针对金相图像的实际，提出了一种新的调焦策略，在较少影响分析效果的同时，能显著减少计算量，提高调焦系统
广东工业大学工学硕士学位论文
的响应速度。
在本文理论的指导下用 Wsual＋＋6．0开发的实验系统，配备有数据库管理，
增强了其实用性，已经应用于深圳某热处理厂，取得了比较好的效果。理论分析
和实际系统上的应用表明，本系统的理论正确，有一定的实际应用价值。
Micrograph sclerometer is a useful tool that used to analyze the hardness of the metal surface and the metal tissues. Presently, there are many kinds of the micrograph sclerometers. And there are some difference between them in the aspects of their work modes and functions. The automatization of the control system and automatically analyzing some metal parameters are the main factors that affect the efficiency and precision of the all system. The current micrograph sclerometers are usually short of them.
Wavelet goes by the name of the microscope of mathematics. It can efficiently locates the time and frequency for the signal. The multi-resolution theory that is a important one in the wavelet transformation theory, and using it we can analyze the signals' approximation and detail in multi-resolution. For the development of the digital image processing techniques and wavelet analysis, we can get more powerful ways to analyze metal parameters with computer. This dissertation primarily study the several important issues with regard to the automation extent and information extent of the measure system of the micrography sclerometers: first, the digital control system of the workbenches; second, the demarcation of the dimension of the metal image; third, automatically measuring the parameters of the metal image using the digital image processing technologies, fourthly, restoring the edges of the test impress fast and efficiently and so resolve the difficult problem that the common digital image processing technol
ogies don't, and finally the automatic focus system. This dissertation lucubrates their difficult problems concerning the theories and the applications. And it have got some innovations in some aspects.
This dissertation analyzes the test impress image in the sample material surface using the multi-resolution theory with some base knowledge about the impress. This dissertation also represents a strategy for analyses and then lucubrates the relevant theories, gets the edges of the impress fast and efficiently. So the system can
automatically analyses the hardness of the region of the material surface. This analyses means is a innovation.
Analyzing the proportion of the second tissue in the whole image region also is a innovation in the application of automatically analyzing the parameters of the metal material surface using the region growth and mathematical morphology. This means is more flexible and exact. In addition, this dissertation also represents another means for analyzing their proportion using the histogram technology, and then compares both in the aspect of their advantage and their application.
This dissertation resolves the demarcation issue for the metal image dimension, and represents the measurement for the diameter of the single crystal or the average diameter of the crystals in the metal micro-tissue.
This dissertation summarizes the current auto-focusing technique, and then represents a new strategy for the metal images. It can remarkably lessen the amount of the data to be computed with less affecting the analyses result. And so the auto-focus system can respond fast.
A experiment system have been developed by Visual C++6.0 according to the relevant theories. In addition, it has a DBMS (Database Management System), and so has more practicability. The system has been applied to a factory in ShenZhen, and has a good effect. It is shown that the theories of this dissertation are right, and have more practice value by the effects of the analyses theoretically and the experiment.
小波变换被称为是数学上的显微镜，具有很好的时、频定位能力。在正交小波变换理论中具有重要地位的多分辨率分析理论，为我们提供了一个多分辨率逼近信号和观察信号细节的工具。小波分析与现代数字图像技术的发展，为计算机分析金相参量提供了强有力的分析工具。本文的主要研究内容就是围绕提高显微硬度计测量系统的信息化、自动化程度中的几个关键问题进行的：一是载物台的数控问题；二是金相图片尺寸的标定；三是利用数字图像处理技术对金相参数进行自动测量；四是对试验压痕边缘进行快速有效地提取，解决了一般数字图像处理技术难以提取其边缘的难题，从而实现硬度的自动计算；最后是自动调焦系统。本文分别从理论和实际应用的角度对其中的难点进行了深入地分析，在几个方面取得了创新。
本文运用基于先验知识的小波多分辨率理论对金相试样上的试验压痕图像进行处理，提出了分析策略，进行了深入的理论研究和实际分析对比，实现对压痕边缘快速而有效的提取，从而实现硬度的自动分析计算，本算法是一个具有特色的创新点。
基于区域生长和数学形态学的第二相相对含量测定技术也是本文在自动分析金相参数应用中具有创新意义的技术，它的灵活性和准确率都比较高；另外，本文还提出了基于直方图技术的金相组织第二相相对含量的测定，并对这两种方法的优劣、应用场合进行了对比分析。
本文解决了金相图片尺寸的标定，提出了对金相组织单个晶粒直径、晶粒平均大小的测量方法。
本文在总结现有自动调焦技术的基础上，针对金相图像的实际，提出了一种新的调焦策略，在较少影响分析效果的同时，能显著减少计算量，提高调焦系统
广东工业大学工学硕士学位论文
的响应速度。
在本文理论的指导下用 Wsual＋＋6．0开发的实验系统，配备有数据库管理，
增强了其实用性，已经应用于深圳某热处理厂，取得了比较好的效果。理论分析
和实际系统上的应用表明，本系统的理论正确，有一定的实际应用价值。
Micrograph sclerometer is a useful tool that used to analyze the hardness of the metal surface and the metal tissues. Presently, there are many kinds of the micrograph sclerometers. And there are some difference between them in the aspects of their work modes and functions. The automatization of the control system and automatically analyzing some metal parameters are the main factors that affect the efficiency and precision of the all system. The current micrograph sclerometers are usually short of them.
Wavelet goes by the name of the microscope of mathematics. It can efficiently locates the time and frequency for the signal. The multi-resolution theory that is a important one in the wavelet transformation theory, and using it we can analyze the signals' approximation and detail in multi-resolution. For the development of the digital image processing techniques and wavelet analysis, we can get more powerful ways to analyze metal parameters with computer. This dissertation primarily study the several important issues with regard to the automation extent and information extent of the measure system of the micrography sclerometers: first, the digital control system of the workbenches; second, the demarcation of the dimension of the metal image; third, automatically measuring the parameters of the metal image using the digital image processing technologies, fourthly, restoring the edges of the test impress fast and efficiently and so resolve the difficult problem that the common digital image processing technol
ogies don't, and finally the automatic focus system. This dissertation lucubrates their difficult problems concerning the theories and the applications. And it have got some innovations in some aspects.
This dissertation analyzes the test impress image in the sample material surface using the multi-resolution theory with some base knowledge about the impress. This dissertation also represents a strategy for analyses and then lucubrates the relevant theories, gets the edges of the impress fast and efficiently. So the system can
automatically analyses the hardness of the region of the material surface. This analyses means is a innovation.
Analyzing the proportion of the second tissue in the whole image region also is a innovation in the application of automatically analyzing the parameters of the metal material surface using the region growth and mathematical morphology. This means is more flexible and exact. In addition, this dissertation also represents another means for analyzing their proportion using the histogram technology, and then compares both in the aspect of their advantage and their application.
This dissertation resolves the demarcation issue for the metal image dimension, and represents the measurement for the diameter of the single crystal or the average diameter of the crystals in the metal micro-tissue.
This dissertation summarizes the current auto-focusing technique, and then represents a new strategy for the metal images. It can remarkably lessen the amount of the data to be computed with less affecting the analyses result. And so the auto-focus system can respond fast.
A experiment system have been developed by Visual C++6.0 according to the relevant theories. In addition, it has a DBMS (Database Management System), and so has more practicability. The system has been applied to a factory in ShenZhen, and has a good effect. It is shown that the theories of this dissertation are right, and have more practice value by the effects of the analyses theoretically and the experiment.
引文
[1]阮秋崎，数字图像处理学，北京：电子工业出版社，2001
[2]郑南宁，计算机视觉与模式识别，北京：国防工业出版社，1998，1～80
[3]陈晋，铸铁金相图像自动分析系统的研究与开发，硕士论文，保存地点：东南大学图书馆，2001
[4]虞伟良，国内外显微硬度计产品发展水平与概况，试验技术及试验机，1998，Vol．38 No．3：3-7
[5]陈武凡，小波分析及其在图像处理中的应用，北京：科学出版社，2002
[6]张兆礼，现代图像处理技术及Matlab实现，北京：人民邮电出版社，2001
[7]Jahne B. Digital Image Processing. Springer, 1997
[8]Cesmeli E., and Wang.D., "Texture Segmentation Using Gaussian-Markov Random Fields and Neural Oscillator Networks", IEEE Transactions on Neural Networks,March 2001,Vol. 12, No.2, pp.394-405
[9]Comer. M.L., and Delp.E.J., "The EM/MPM Algorithm for Segmentation of Textured Images: Analysis and Further Experimental Results", IEEE Transactions on Image processing. October 2000, Vol.9, No.10, pp.1731-1744.
[10]Strickland. R.N, "Wavelet transform methods for image detection and recovery",Transactions on image processing, 1997,Vol.6.pp.724-734.
[11]Hsin.H., "Texture Segmentation Using Modulated Wavelet Transform", IEEE Transactions on Image Processing, July 2000,Vol.9, No.7, pp. 1299-1303
[12]Randen.T., and husoy, J.H., "Texture Segmentation using Filters with Optimized Energy Separation", IEEE Transactions on Image Processing, April 1999, Vol.8, No.4, pp. 571-583.
[13]马颂德，张正友．计算机视觉-计算理论和算法基础，北京：科学出版社，1998．
[14]何东健，杨青，实用图像处理技术，西安：陕西科技出版社，1998，30～65
[15]Castleman K.R.数字图像处理，朱志刚等译，北京：电子工业出版社，1998
[16]夏德深，傅德胜．现代图像处理技术及应用．南京：东南大学出版社，1997
[17]赵荣椿等．数字图像处理导论，西安：西北工业大学出版社，1995
[18]章毓晋，图象处理和分析，北京：清华大学出版社，1998
[19]崔屹，数字图象处理技术与应用，北京：电子工业出版社．1997
[20]高文，计算机视觉一算法与系统原理．北京：清华大学出版社，1999
[21]吴立德，计算机视觉．上海：复旦大学出版社．1993
[22]程存学，朱晓昆．计算机视觉—低层处理技术，北京：电子工业出版社，1993
[23]崔锦泰，小波分析导论．西安：西安交通大学出版社，1995
[24]徐佩霞，孙功宪，小波分析与应用实例，合肥：中国科技大学出版社，1996
[25]刘贵忠，小波分析及其应用．西安：电子科技大学出版社，1992
[26][法]Y.迈耶，小波分析与信号处理—理论、应用及软件实现，尤众译，北京：世界图书出版公司，1992
[27]杨福生．小波变换的工程分析与应用．北京：科学出版社，1999
[28]付忠良，图像阈值选取方法的构造，中国图形图像学报，2000，5(6)．
[29]余新平，朱立，一种具有抗噪声干扰的图像边缘提取算法的研究，电子技术应用，25(1)，1999：9～10
[30]Hwang, W.1.Chun-Shien lu.Pau.Choo Chung." Shape from texture: estimation of planar surface orientation through the ridge surfaces of continuous wavelet transform", IEEE Transactions on Pattern Analysis and Machine Intelligence, Nov. 1995,Vol. 17, No. 11.pp. 1043-1056
[31]Jain. A.K, and Farrokhnia.F. "Unsupervised texture segmentation using Gabor filter." Pattern Recognition, 1991,Vol.24.pp. 1167-1185,
[32]Dunn.D. and Higgins. W.E. "Optimal Gabor filters for texture segmentation". IEEE Transactions on image processing, 1995,Vol.4.pp.947-964.
[33]Unser. M. "Texture classification and segmentation using wavelet frames", IEEE Transactions on image processing, 1995,Vol.4.pp. 1549-1560.
[34]Kasparis.T, Tzannes.N.S., Bassiouni.M., and Chen. Q., "Texture Description based on Fractal and Energy Features", Computers and Electrical Engineering, 1995,Vol.21.pp.21-32
[35]R.Adams and L.Bischof. "Seeded region growing". IEEE Transactions on Pattern Analysis and Machine Intelligence. 1994,16(6):641-647
[36] Mallat. S., "A Wavelet Tour of Signal Processing". Academic Press. San Diego,CA, 1998.
[37] Vetterli,M., and Herley. C., "Wavelets and Filter Banks: Theory and Design", IEEE Trans. Signal Processing, Sept. 1992,Vol.40, No.9, pp.2207-2232.
[38] Priestley. M.B., "Wavelets and Time-Dependent Spectral Analysis", Journal of Time Series Analysis. 1996, Vol.17, No.Ⅰ, pp.85-103
[39] Canny J, "A computational approach to edge detection", T-PAMI. 1998, 8(6):679-698
[40] Bergholm F, "Edge focusing", T-PAMI. 1987,9(6):726-741
[41] Zhang Y.J, "A Survey on evaluation methods for image segmentation", Pattern Recognition, 1996,29(8): 1335-1346
[42] Sahoo P, Wuukins, Yenger J, "Threshold Selection Using Renyi's Entropy", Pattern Recognition, 1997,30(1):394-398
[43] Kmizinga P., and Petkov. N., "Nonlinear Operator for Oriented Texture", IEEE Transactions on Image Processing, October 1999, Vol.8, No.10. pp.1395-1408.
[44] 李现勇，Visual C++串口通信技术与工程实践，北京：人民邮电出版社，2002
[45] 清汉计算机工作室，Visual C++6.0多媒体开发实例，北京：机械工业出版社，2000
[46] 盛，李见为，基于智能化自动调焦的高级显微镜系统，光电工程，2000，Vol．27，33-36
[47] 白立芬，基于图像处理的显微镜自动调焦方法研究，仪器仪表学报，1999，Vol．20．6，612-614
[48]郭彦珍，图像测量技术中一种调焦的判别方法，西安理工大学学报，2001，Vol．17．1，40-42
[49] 王任华，沈忙作，自动对焦算法研究，光电工程，2000，Vol．27．4，11-13
[50] H.O.Lim, W. Zhang and L.M.Koh. "Automated visual inspection for IC wire-bond using auto-focusing technigue", IEEE Electronics Manufacturing Technology Symposium, 1993, 31-36
[2]郑南宁，计算机视觉与模式识别，北京：国防工业出版社，1998，1～80
[3]陈晋，铸铁金相图像自动分析系统的研究与开发，硕士论文，保存地点：东南大学图书馆，2001
[4]虞伟良，国内外显微硬度计产品发展水平与概况，试验技术及试验机，1998，Vol．38 No．3：3-7
[5]陈武凡，小波分析及其在图像处理中的应用，北京：科学出版社，2002
[6]张兆礼，现代图像处理技术及Matlab实现，北京：人民邮电出版社，2001
[7]Jahne B. Digital Image Processing. Springer, 1997
[8]Cesmeli E., and Wang.D., "Texture Segmentation Using Gaussian-Markov Random Fields and Neural Oscillator Networks", IEEE Transactions on Neural Networks,March 2001,Vol. 12, No.2, pp.394-405
[9]Comer. M.L., and Delp.E.J., "The EM/MPM Algorithm for Segmentation of Textured Images: Analysis and Further Experimental Results", IEEE Transactions on Image processing. October 2000, Vol.9, No.10, pp.1731-1744.
[10]Strickland. R.N, "Wavelet transform methods for image detection and recovery",Transactions on image processing, 1997,Vol.6.pp.724-734.
[11]Hsin.H., "Texture Segmentation Using Modulated Wavelet Transform", IEEE Transactions on Image Processing, July 2000,Vol.9, No.7, pp. 1299-1303
[12]Randen.T., and husoy, J.H., "Texture Segmentation using Filters with Optimized Energy Separation", IEEE Transactions on Image Processing, April 1999, Vol.8, No.4, pp. 571-583.
[13]马颂德，张正友．计算机视觉-计算理论和算法基础，北京：科学出版社，1998．
[14]何东健，杨青，实用图像处理技术，西安：陕西科技出版社，1998，30～65
[15]Castleman K.R.数字图像处理，朱志刚等译，北京：电子工业出版社，1998
[16]夏德深，傅德胜．现代图像处理技术及应用．南京：东南大学出版社，1997
[17]赵荣椿等．数字图像处理导论，西安：西北工业大学出版社，1995
[18]章毓晋，图象处理和分析，北京：清华大学出版社，1998
[19]崔屹，数字图象处理技术与应用，北京：电子工业出版社．1997
[20]高文，计算机视觉一算法与系统原理．北京：清华大学出版社，1999
[21]吴立德，计算机视觉．上海：复旦大学出版社．1993
[22]程存学，朱晓昆．计算机视觉—低层处理技术，北京：电子工业出版社，1993
[23]崔锦泰，小波分析导论．西安：西安交通大学出版社，1995
[24]徐佩霞，孙功宪，小波分析与应用实例，合肥：中国科技大学出版社，1996
[25]刘贵忠，小波分析及其应用．西安：电子科技大学出版社，1992
[26][法]Y.迈耶，小波分析与信号处理—理论、应用及软件实现，尤众译，北京：世界图书出版公司，1992
[27]杨福生．小波变换的工程分析与应用．北京：科学出版社，1999
[28]付忠良，图像阈值选取方法的构造，中国图形图像学报，2000，5(6)．
[29]余新平，朱立，一种具有抗噪声干扰的图像边缘提取算法的研究，电子技术应用，25(1)，1999：9～10
[30]Hwang, W.1.Chun-Shien lu.Pau.Choo Chung." Shape from texture: estimation of planar surface orientation through the ridge surfaces of continuous wavelet transform", IEEE Transactions on Pattern Analysis and Machine Intelligence, Nov. 1995,Vol. 17, No. 11.pp. 1043-1056
[31]Jain. A.K, and Farrokhnia.F. "Unsupervised texture segmentation using Gabor filter." Pattern Recognition, 1991,Vol.24.pp. 1167-1185,
[32]Dunn.D. and Higgins. W.E. "Optimal Gabor filters for texture segmentation". IEEE Transactions on image processing, 1995,Vol.4.pp.947-964.
[33]Unser. M. "Texture classification and segmentation using wavelet frames", IEEE Transactions on image processing, 1995,Vol.4.pp. 1549-1560.
[34]Kasparis.T, Tzannes.N.S., Bassiouni.M., and Chen. Q., "Texture Description based on Fractal and Energy Features", Computers and Electrical Engineering, 1995,Vol.21.pp.21-32
[35]R.Adams and L.Bischof. "Seeded region growing". IEEE Transactions on Pattern Analysis and Machine Intelligence. 1994,16(6):641-647
[36] Mallat. S., "A Wavelet Tour of Signal Processing". Academic Press. San Diego,CA, 1998.
[37] Vetterli,M., and Herley. C., "Wavelets and Filter Banks: Theory and Design", IEEE Trans. Signal Processing, Sept. 1992,Vol.40, No.9, pp.2207-2232.
[38] Priestley. M.B., "Wavelets and Time-Dependent Spectral Analysis", Journal of Time Series Analysis. 1996, Vol.17, No.Ⅰ, pp.85-103
[39] Canny J, "A computational approach to edge detection", T-PAMI. 1998, 8(6):679-698
[40] Bergholm F, "Edge focusing", T-PAMI. 1987,9(6):726-741
[41] Zhang Y.J, "A Survey on evaluation methods for image segmentation", Pattern Recognition, 1996,29(8): 1335-1346
[42] Sahoo P, Wuukins, Yenger J, "Threshold Selection Using Renyi's Entropy", Pattern Recognition, 1997,30(1):394-398
[43] Kmizinga P., and Petkov. N., "Nonlinear Operator for Oriented Texture", IEEE Transactions on Image Processing, October 1999, Vol.8, No.10. pp.1395-1408.
[44] 李现勇，Visual C++串口通信技术与工程实践，北京：人民邮电出版社，2002
[45] 清汉计算机工作室，Visual C++6.0多媒体开发实例，北京：机械工业出版社，2000
[46] 盛，李见为，基于智能化自动调焦的高级显微镜系统，光电工程，2000，Vol．27，33-36
[47] 白立芬，基于图像处理的显微镜自动调焦方法研究，仪器仪表学报，1999，Vol．20．6，612-614
[48]郭彦珍，图像测量技术中一种调焦的判别方法，西安理工大学学报，2001，Vol．17．1，40-42
[49] 王任华，沈忙作，自动对焦算法研究，光电工程，2000，Vol．27．4，11-13
[50] H.O.Lim, W. Zhang and L.M.Koh. "Automated visual inspection for IC wire-bond using auto-focusing technigue", IEEE Electronics Manufacturing Technology Symposium, 1993, 31-36