摘要
论文标题中提及到的土石,一般是由作为骨料的砾石或卵石与作为填充料的粘土或砂土组成的松散体,它是介于岩体和土体两者之间的一种特殊地质体。油新华博士把它称之为土石混合体,由于其具有特殊的物理力学特性,因而它是土木工程中一个必需认真研究的课题。
自分形几何创立以来,技术人员开始用粗料颗粒分布分维、粗料轮廓分维这些参数指标作为土石混合体微结构研究的依据,并将其与宏观物理力学性质相联系,以获得它们之间的定量关系,进而解释土石混合体的力学行为特性,并以这些参数作为土石混合体工程分类的依据。
但目前技术人员只能以纯手工操作的筛分析实验来计算粒度分维数,从而展开对土石混合体微结构研究。该方法准确有效,但适用范围窄,操作不便,极大地制约了工程人员对土石混合体微结构的研究。
本文通过大量实验,尝试性地实现了一种利用数字图像处理技术来测量土石混合体分形维数的检测流程,并提出了一种改进的基于二值图的8向跟踪封闭边缘提取算法。
本文首先通过实验比较了常用的去噪算法的效果之后得到一种有效的土石混合体去噪算法。随后又以大量实验的方法比对了常用的分割算法并将图像二值化。在二值图的基础上,比对了众多常用的边缘检测算法效果之后,改进了一种基于二值图的8向跟踪封闭边缘提取算法。该算法同时完成封闭边缘提取、降噪以及孔洞的填充,同时为后面周长和面积计算的准确性提供了保障。
本文以数字图像技术完成了土石混合体分形维数的检测,减轻了技术人员手工检测的劳动强度,丰富了工程人员开展土石混合体微结构研究的手段,扩大了分形维数的适用范围。同时对基于数字图像处理技术的在其它领域内的分形检测也具有实用参考价值。
The "earth-rock" in the title of this paper is a loose rock which is composed of gravel or cobble which used as aggregate and clay or sand which used as filling material. Doctor Xinhua Liu called it as earth-rock aggregate and classified it between rock mass and soil mass. For its special characteristics of physical mechanical, the earth-rock is a import study subject in civil engineering domain.
From the establishment of fractal geometry, technician had started to use some parameters including Distribution fractal dimension of coarse grain and coarse grain boundary contour line as a new research method on microstructure of earth-rock aggregate. They combined the parameters and macroscopic properties of physical mechanical of earth-rock aggregate to get their quantitative relation which be used to explain the characteristics of mechanical behavior of earth-rock aggregate. The parameters of distribution fractal dimension can also be classification basis for engineering of earth-rock aggregate.
Now technician can only calculate grain-size fractal dimension by sieve analysis in manual operation way to research the microstructure of earth-rock aggregate. Although the way of sieve analysis is accurate and effective, it restricted the technician to research further for its narrow of scope of application and its inconvenient operation.
In this paper realized experimentally one process of distribution fractal dimension detection of earth-rock aggregate base on digital image process through a lot of experimental and modify an algorithm of enclosed edge extraction base on the algorithm of 8-direction edge tracing in binary image.
At first, this paper get a more effective de-noising algorithm after comparing the result of de-noising algorithm of commonly used. Subsequently, this paper got binary image when found a optimum segmentation algorithm. Later, having compared the result of conventional algorithm of edge detection, this paper modified an algorithm of enclosed edge extraction which accomplish enclosed edge extraction, de-noising and hole-filling in same time and guarantee the accurate of calculating perimeter and area.
The Detection of distribution fractal dimension of earth-rock aggregate base on digital image process decreased labor intensity for technician to detect the distribution fractal dimension in manual operation and enriched the research method of microstructure of earth-rock aggregate for technician and enlarged the scope of application for fractal and use for reference to some detection of distribution fractal dimension base on digital image process in other domain.
引文
[1]油新华,土石混合体随机结构模型及其应用研究[博士学位论文],北方交通大学,2002
[2]Mandelbrot B B,The fractal geometry ofnature[M],上海:上海远东出版社,1998
[3]冯志刚,周宏伟,图像的分形维数计算方法及其应用[J],江苏理工大学学报,(2001)22(6).92-95
[4]叶瑞英,马红,使用分形维数对超高强度钢断裂韧性的研究[J],四川大学学报(自然科学版),(2001)38(2),210-214
[5]彭瑞东,谢和平,鞠杨,二维数字图像分形维数的计算方法[J],中国矿业大学学报,(2004)33(1),19-24
[6]郭飞,徐绍辉,刘建立,土壤样本分形几何特征的图像分析方法[J],土壤学报,(2005)42(1).24-28
[7]晏磊,罗立,张雪虎,真实孔径雷达海洋图像的分形特征分析[J],电波科学学报,(2007)22(4),604-609
[8]李国宾,孟歆,关德林,魏海军,基于小波和分形提取磨粒图像特征参数的研究[J],内燃机学报,(2006)24(5),476-479
[9]邹群彩,凌祥,涂善东等,基于分形理论的岩相图像识别与分析[J],南京化工大学学报,(2000)22(1),28-31
[10]李志铭等,定量金相分析的图像技术[J],光电工程,(1997)22(4),56-59
[11]Mandelbrot,B.B.,Les Objects Fractals:form,hazard et dimention[M],Pairs,Flammarion,1975
[12]谢和平,岩土介质的分形孔隙和分形粒子[J],力学进展,(1993)24(10),145-164
[13]郭飞,徐绍辉,刘建立,土壤样本分形几何特征的图像分析方法[J],土壤学报,(2005),42(1),24-28
[14]张济忠,分形[M],北京:清华大学出版社,1995
[15]Feder,J.fractals[M],New York,Plenum Press,1988
[16]董连科,分形与应用[M],辽宁:辽宁科学出版社,1992
[17]谢和平,分形-岩石力学导论[M],北京:科学出版社,1996
[18]张济忠,分形[M],北京:清华大学出版社,1995
[19]Falconer K.Fractal Geometry,分形几何[M],曾文曲,刘世耀译,沈阳:东北工业出版社,1991
[20]黄涛,数字图像的平滑[J],韶关学院学报,(2004)12(06),13-15
[21]朱秀昌,刘峰,胡栋,数字图像处理与图像通信[M],北京:北京邮电大学出版社,2002
[22]郎锐,数字图像处理VisualC++实现[M],北京:北京希望电子出版社,2002
[23]杨静宇,曹雨龙,计算机图像处理及常用算法手册[M],南京:南京大学出版社,1997
[24]章毓晋,图像处理和分析基础[M],北京:高等教育出版社,2002,106-116.
[25]赖志国,余啸海,Matlab图像处理与应用[M],北京:国防工业出版社,2004
[26]贾永红,计算机图像处理与分析[M],武汉:武汉大学出版社,2001
[27]刘文耀,数字图像采集与处理[M],电子工业出版社,2007
[28]刘文萍,吴立德,图像分割中阈值选取方法比较研究[J],模式识别与人工智能,(1997)3(3),35-38
[29]章毓晋,图像分割[M],清华大学出版社,2000
[30]Babaguchi N,Yamada K,Kisc K,Tezuka T,Connectionist Model Binarization.In:Proc of the 10th International Conference on Pattern Recognition,New Jersey,USA,1990,S1-S6
[31]Ghosh A,Pal N R,Pal S K.Image Segmentation Using a Neural Network.Biological Cybernetics,(1991)66(2),151-158
[32]Shah J.Parameter Estimation,Mutiscale Representation and Algorithms for Energy Minimizing Segmentation.In:Proc of the 10th International Conferenc on Pattern Recognition,New Jersey USA,(1990),815-819
[33]Murthy C A,Pal S K.Fuzzy Thresholding:Mathematical Framwork,Bound Functions and Weighted Moving Average Techniques.Pattern Recognition Letter,(1990)11(3),197-206
[34]Dinstein Ac Fong,Fast Distribution Between Homogeneous and Textured Region[J],Proceedings of 7th Conference on PR,1984
[35]Chanda B,Kundu M K,Padmaha Y V,A Multi-Scale Morphologic Edge Detector,Pattern Recognition,(1998)31(10),1469-1478
[36]田捷,实用图像分析与处理技术[M],电子工业出版社,1995
[37]K.R.Castleman,数字图像处理[M],电子工业出版社,1998
[38]王建萍,王家平,数字图象处理在定量金相分析中的应用[J],材料导报,(2003)17(1),58-61
[39]J.Prewitt,Object Enhancement and Extraction,Picture Process,Psychopict,(1970),75-149,
[40]L.Cx Roberts,Machine Perception of Three-Dimension Solids.in Optical and Electro-Optimal Information Processing,J.t.Tippett,etal.,Ed.Cambridge,MA:MTT Press,(1965),99-197
[41]R.A.Kirsch,Computer Determination of the constituent structure of Biological images,Computer and Biomedical research,(1971)1(18),113-125
[42]L.S.Davis,A Survey of Edge Detection Techniques,CGIP,(1975)4,248-270,
[43]L.Sobel,Camera Models and Machine Perception,PhD theses,Stanford University,Standford,CA,1970
[44]A.L.Yuille,T.Poggio,Fingerprint theorems for zero crossing[M],Opt.Soc.Am.A,(1985)2,683-692
[45]王新成,高级图像处理技术[M],北京:中国科技出版社,2001
[46]崔凤魁,定量金相分析算法及实现[J],材料科学与工艺,(2000)8(3),109-112
[47]赵荣椿,数字图像处理导论[M],西安:西北工业大学出版社,1995
[48]崔凤魁,张丰收,白露等,二值图像目标邻域点法边界跟踪算法[J],洛阳工学院学报,(2001)22(1),28-34
[49]柳稼航,杨建峰,单新建等,一种基于优先搜索方向的边界跟踪算法[J],遥感技术与应用,(2004)19,209-213
[50]王福生,齐国清,二值图像中目标物体轮廓的边界跟踪算法[J],大连海事大学学报,(2006)1,62-64
[51]钟志光,卢君,刘伟荣,Visual C++.NET数字图像处理实例与解析[M],北京:清华大学出版社,2003
[52]Donald H,Pauline B M,计算机图形学[M],北京:清华大学出版社,1998
[53]Roger D F,计算机图形学编程基础[M],北京:中国机械工业出版社,2002
[54]Dunlavey M R,Efficient Polygon-filling Algorithm for Raster Displays[J],ACM Transactions on Graphics,(1983)2(4),264-273.
[55]Guy Eddon,Henry Eddon,组件编程技术(希望图书创作室)[M],希望电子出版社,1998
[56]王开平,游志胜,高阳,二维数字化目标周长的精确计算[M],中国图像图形学报,(1997)2(3),165-170
[57]Kulap Z,Area and perimeter measurement of blobs in discrete binary picture[J],Computer Graphics Image Processing,(1977),13(6),434-451
[58]Kiyati N,Mayday D,Calculating geometric properties from Fourier representation[J],Pattern Recognition,(1989)32(5),469-475
[59]朱长青,杨绪华,邱振戈,数字图像区域周长计算的原理和方法[J],测绘工程,(1999)8(2),29-33
[60]吴小培,孔敏,张德龙,一种计算封闭区域周长和面积的新方法[J],数据采集与处理,(1999),14(2),244-246