摘要
由于目前先进设计制造技术的需要,零件的三维模型的作用越来越重要,对于目前已存在的零件图纸,由工程图纸到三维模型的过程需要企业花费大量的人力去完成。机械零件图的自动识别及三维模型成型技术就是将机械零件工程图纸通过矢量化方法提取出有用图元特征,保存为计算机辅助设计制造软件能识别的通用矢量格式DXF,并在Pro/E二次开发的基础上快速准确地生成模型的过程。
本文主要研究回转体机械零件图的自动识别及三维成型技术,深入分析了现有矢量化技术及Pro/E二次开发技术发展情况,提出了基于Pro/TOOLKIT应用程序的工程图纸到零件三维模型成型的方法,首先通过二值化、综合形态学去噪、中值滤波、细化等方法得到图像中成型三维模型最重要的轮廓信息,利用Hough变换检测出线段及圆弧,在线段端点及长度的确定过程中结合π/2线,不仅提高了线段检测的运算速度,而且降低了碎线段发生的概率,识别图形的过程中表现出的良好鲁棒性及线段的正规性是后续的零件三维模型生成的良好基础,对圆弧的识别中应用了随机Hough变换,有效减少了程序运行的时间。系统还完成了对识别出的图形编辑修改及DXF文件的保存。
零件的三维模型成型主要利用了Pro/E三维软件的模型成型优势,对识别出的矢量文件打开实现回转体特征的成型,在零件细节特征上,本文制作了回转体零件的细节特征用户自定义特征库,使用的过程中选择不同特征放置于模型上即可,通过设定自定义特征的可变尺寸可实现不同尺寸特征的放置。
最后本文研究了基于Pro/TOOLKIT二次开发的参数化设计方法,以已存在的零件模型为样板,实现了不同零件的同一系统的参数化设计,可以很大程度地提高复杂零件的设计效率。
At present, because of the need of advanced designing and manufacturing technology, three-dimensional model of part play a more significant role. For existented engineer drawings, it’s a waste of energy for company to convert it to models directly. Research on automatically identifying of revolving parts’engineering drawing and forming of three-dimensional model is save engineer drawing in the format of DXF by vectorizing the drawing in order to identify useful feature and then form the three-dimensional model quicky and righty based on the secondry research of Pro/E.
In this paper, the research faces to revolving parts, offers the method of forming three-dimensional model from drawing based on Pro/TOOLKIT applications after analysed the development of vectorization and secondry research of Pro/E techniques thoroughly. The first process is processing the drawing image, including image binarizing, noise-reducing, smoothing, and thinning, in order to get the important outline of the part to form the model. And then detect the line, circle and arc by Hough transformation. In the process of fitting a line,π/2 line is used for the end point confirming, not only improved the speed of the motion, but also reduced the probability of fragmentary line. In the process of identifying, the strong robustness and normality is a good base for three-dimensional model’s forming. In the process of recognizing arc, Randomized Hough Transform is used, it can improve the speed of program. The system also include edit, modify and DXF file’s save of the drawing recognized.
Three-dimensional model’s forming takes advantage of Pro/E in forming, open the vector Files in Pro/E and form the model. For the details of feature, user-defined feature library is made in this paper. In the process of using, user can select the feature needed to put on the rovelving part. It can put feature in different dimension if alterable dimension is set in the user-defined feature founding.
At last, the parameterized design method based on the second research is researched. It uses an existing model as a sample plate, realized different parts’parameterized design in a same system. It’s a great help of the speed of complex parts’design.
引文
[1]肖立峰,毕树生,刘静华,等.工程图矢量化在机械设计教学中的应用研究与探讨,全国机械设计教学研讨会议(2006):210~214.
[2]龚洁晖,张慧.基于工程图的三维重建研究.软件学报,2008,19(7):1794~1805.
[3]李四明.工程图纸输入与自动识别系统的研究,[硕士学位论文].北京:中国农业大学,2000.
[4]张明君.基于位图的自动矢量化研究,[硕士学位论文].西安:西安电子科技大学,2004.
[5] Dori D, Tombre K. From engineering drawings to 3D CAD models: Are we ready now? Computer-Aided Design, 1995, 27(4): 243~254.
[6]Hubbard C, Kim YS.Geometric assistance for visual reasoning and construction of solids with curved surfaces from 2D orthographic views. Advances in Engineering Software, 2001, 32(1):21~35.
[7]韩东.三维重构中的匹配技术研究,[硕士论文].大连:大连理工大学.2006.
[8]秦斌杰,陈旭.医学图像三维可视化.航天医学与医学工程,2001,14(6):452~455.
[9]Masuda,H.Numao,M.A cell-based approach for generating solid objects from orthographic projections Computer-Aided Design 1997,29(3),177~187.
[10]Chen Z,Perng D. Automatic reconstruction of 3-D solid ojects 2-D orthographic views Pattern Recognition 1988, 21:439~449.
[11]You.C. and Yang.S Reconstruction of curvilinear manifold objects from orthographic views Computer & Graphics 1996,20(2),275~293
[12]Kim,C.Inoue,M;Nishihara,S .Heuristic understanding of three orthographic vilews Journa1 of Information Processing 1992,15(4),510~518.
[13]Watanabe T.Tashiro A.Fujii S. Estimation of three-dimensional objects from orthographic views with inconsistencies Computer & Graphics, 1995, 19(6), 815~829.
[14]Simon S.P.Shum,W,S,Iau,Matthew M,F.Yuen and K.M.Yu. Solid reconstruction from orthographic opaque views using incremental extrusion. Computer & Graphics, 1997, 21(6), 787~800.
[15]Jan Ganesant and Venkat Devarajan Intersecting features extraction from orthographic projections. Computer-Aided Design, 1998, 30(11), 863~873.
[16]X Xu and S Hinduja Recognition of rough machining features in 21/2D components. Computer-Aided Design, 1998, 30(7), 503~516.
[17]Byeong-Seok Shin and Yeong Gikl Shin. Fast 3D solid model reconstruction from orthographicviews. Computer-Aided Design, 1998, 30(1), 63~67.
[18]Chang-Hum Kim Beom-Soo Oh Woong-Soon Kim Atea-Based Reconstruction of 3D Curvilnear Objects from Two Orthographic Views Proceedings of CAD & Graphics'97 The international conference on CAD & CG 1997.
[19]陆国栋.基于工程语义的三维重建方法研究,[博士学位论文].杭州:浙江大学,1999.
[20]Wang W, Grinstein GG. A survey of 3D solid reconstruction from 2D projection line drawings. Computer Graphics Forum, 1993, 12(2):137~158.
[21]Idesawa M. A system to generate a solid figure from three view. Bull. JSME., 1973,16:216~225.
[22]Markowsky G, Wesley MA. Fleshing out wireframes. IBM Journal of Research and Development, 1980, 24(5):582~597.
[23]Gujar UG, Nagendra IV. Construction of 3D solid objects from orthographic views. Computer & Graphics, 1989, 13(4):505~521.
[24]Yan QW, Chen CLP, Tang ZS. Efficient algorithm for the reconstruction of 3D objects from orthographic projections. Computer-Aided Design, 1994, 26(9):699~717.
[25]Preiss K. Algorithms for automatic conversion of a 3-view drawing of a plane-faced part to the 3-D representation.Computers in Industry, 1981, 2(2):133~139.
[26] Sakura H,Gossard D C. Solid model input through orthographic views. Computer Graphics, 1983, 17(1):243~247.
[27]Lequette R. Automatic construction of curvilinear solids from wire-frame views.Computer-Aided Design, 1988, 20(4):171~180.
[28]辜凯宁,唐泽圣,孙家广.从三视图重建三维形体技术的研究.计算机学报,1986, 11(6):401~409.
[29] Aldefeld B. On automatic recognition of 3-D structures from 2-D representations. Computer-Aided Design, 1983, 15(2):59~64.
[30] Nagasamy V, Langrana NA. Reconstruction of three-dimensional objects using a knowledge-based environment.Engineering with Computers, 1991, 7(1):23~35.
[31]Ho B. Inputting constructive solid geometry representations directly from 2D drawings. Computer-Aided Design, 1986, 18(3):147~155.
[32]Dimri J, Gurumoorthy B. Handling sectional views in volume-based approach to automatically construct 3D solid from 2D views. Computer-Aided Design, 2005, 37(5):484~495.
[33]Cicek A, Gulesin M. Reconstruction of 3D models from 2D orthographic views using solid extrusion and revolution. Journal of Materials Processing Technology.2004, 152:291~298.
[34]Lee H, Han S. Reconstruction of 3D interacting solids of revolutioe from 2D orthographic views. Computer-Aided Design, 2005, 37(13):1388~1398.
[35]公茂凯,高国安,石森.由三视图构造三维实体方法的综述.计算机研究与发展,1992,(8):47~52.
[36]Tombre K. Analysis of engineering drawing: state of the art and challenge .In: Tombre k, Chhabra A K(eds).Graphics Recognition-Algorithms and systems.Vol.1389 of Lecture Notes in Computer Science. Springer-Verlag, Berlin, Germany.1998:257~264.
[37]Lam L, Lee S W, Suen C Y. Thinning methodologies–a comprehensive survey.IEEE Trans. Pattern Analysis and Machine Intelligence, 1992, 14(9):869~885.
[38]Shih F Y, Pu C C.A skeletonization algorithm by maxima tracking on euclidean distance transform [J].Pattern Recognition, 1995, 28(3):331~341.
[39]Golland P, Eric W, Grimson L. Fixed topology skeletons[C]//IEEE, ed. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Vol 1.Los Alamitios, CA: IEEE Computer Society,2000:10~17.
[40]Lin X, Shimotsuji S, Minoh M, et al. Efficient diagram understanding with characteristic pattern detection[J].Computer Vision, Graphics and Image Processing,1985,30:84~106.
[41]Song Jiqiang, Su Feng, Chiew-Lan Tai, et al. An object-oriented progressive-simplification based vectorization system for engineering drawings: model, algorithm and performance.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(8):1048~1060.
[42]郭丙炎,朱建新,周济.工程图形扫描图像的智能识别.计算机应用研究,1998(3):41~42.
[43]王金鹤,欧宗瑛,夏晓东.工程扫描图象的直线整体识别算法.中国图象图形学报,1998,3(11):912~917.
[44]席晓鹏,糜宁芳,罗志伟,等.工程图中的模板识别和匹配方法.计算机应用研究,2003,12:61~64.
[45]易成贵,王云航,谭剑.工程图自动识别中自适应块算法的研究.兵工自动化,1997,4:1~4.
[46]宋晓宇,王永会.矢量化算法的研究[J].计算机工程与应用,2000,3:41~43.
[47]路通,席晓鹏,苗明,等.建筑工程图识别与理解—模型与算法.计算机研究与发展,2005,42(1):144~152.
[48]潘华伟,邹北骥,李莉,等.基于图像的工程图视图分离.计算机工程,2007,33(18):276~278.
[49]李学营,苏洲.直线及圆弧矢量化方法的研究.佳木斯大学报,2007,25(1):114~119.
[50]廖友军,王恒升,余金伟.轴类零件二维工程图的特征识别方法.工程图学报,2007,2:174~178.
[51]秦红斌,刘云华,张卫国,等.三牙轮钻头牙轮形状特征识别技术研究.组合机床与自动化加工技术,2008,9:24~29.
[52]刘玲,周旭东.Visual C++6.0对Pro/ENGINEER 2001二次开发参数化的研究.机械,2004, 31(9):39~41.
[53]徐慧娟,夏治,谢静.PRO/E的应用现状.中国制造业信息化,2005,34(4):92~93.
[54]乔小燕,傅炅,王军.基于Pro/E二次开发的冲模零件参数化设计.模具工业,2007,33(12):8~14.
[55]李占奎.基于Pro/E二次开发技术在乳化液泵站生产建模中的应用.煤矿机械,2008,29(3):179~180.
[56]谢敏理,和丽,许纪倩.Pro/E二次开发实现自动装配的一种方法.现代机械,2006,1:46~48.
[57]吴斌.Pro/E装配干涉检查的二次开发,[硕士论文].南京:东南大学,2006.
[58]曹云鹏.基于Pro/E二次开发技术的凸轮零件设计,[硕士论学文].北京:华北电力大学,2006.
[59]张友湖,夏超群.基于Pro/ENGINEER Wildfire的NC加工二次开发.制造技术与机床,2008(6):118~120.
[60]高磊,王军英,王挺昊,等.以Pro/E产品结构树为基础的CAPP.机械设计与制造,2001(4):33~35.
[61]孙江宏,刘博.Pro/ENGINEER Wildfire高级应用详解.北京:中国水利水电出版社,134~137.
[62]丁刚.基于Pro/Program方法及族表的产品参数化设计.机械工程师,2008(5):51~52.
[63]二代龙震工作室.Pro/TOOLKIT Wildfire2.0插件设计.北京:电子工业出版社,2005:44~59.
[64]蒋知民,张洪穗.机械图样新旧国家标准对照汇编.北京:国防工业出版社,1996:8~10.
[65]江明,刘辉,黄欢.图像二值化技术的研究.软件导刊,2009,8(4):175~177.
[66]陈刚,王厚大.一种基于形态学的文本图像二值化算法.信息化纵横,2009,16:60~62.
[67]杨淑莹.VC++图像处理程序设计.北京:清华大学出版社,2005:188~205.
[68]领域滤波.图像细化的VC实现.科技信息,2008(16):376~377.
[69]中值滤波.图像细化的VC实现.科技信息,2008(16):376~377.
[70]赵耀,陈慧女.图像细化的VC实现.科技信息,2008(16):376~377.
[71]Hori O, Tanigawa S. Raster-to-vector conversion by line fitting based on contours and skeletons [C]//Proceedings of the 2nd International Conference on Document Analysis and Recognition, Tsukuba,Japan,1993:623~626.
[72]Vaxiviere P, Tombre K. Subsampling: a structural approach to technical document vectorization [C]//Dori D, Bruckstein A (eds). Shape, Structure and Pattern Recognition (Proceedings of the IAPR Workshop on Syntactic and Structural Pattern Recognition,Nahariya Israel, 1994). World Scientific,1995:323~332.
[73]宋继强,苏丰,陈冀兵,等.基于种子段的方向无关的直线矢量化方法.软件学报,2000, 11(9):1196~1201.
[74]谭建荣,彭群生.基于图形约束的工程图纸扫描图像直线整体识别方法.计算机学报, 1994, 17(8):561~569.
[75]赵洪亮.基于Hough变换改进的直线检测与定位.激光与红外,2009,39(9):1009~1011.
[76]姬忠校.机械零件图纸的矢量化及拼接技术研究,[硕士学位论文].西安:西安建筑科技大学,2008.
[77]Duda R O, Hart P E. Use of the Hough transformation to detect lines and curves in pictures. Communications of the ACM,1972,15(1):11~15.
[78]杨四海,陈锻生,谢维波.Hough变换的特性分析:一种全局观点.计算机辅助设计与图形学学报,2006,18(8):1197~1204.
[79]Xu L, qa E, Kultanen P.Randomized Hough Transform (RHT):Basic Mechanisms, Algorithms, and Computational Complexities[J].Comput Vision Graphics Image Process: Image Understanding, 1993,57(2):131~154.
[80]H. K. Yuen,J. Princen,J. Illingworth,J. Kittler. Comparative study of Hough transform methods for circle finding. Image and Vision Computing, 2004, 8(1):71~77.
[81]陈建春.矢量图形系统开发与编程.北京:电子工业出版社,2006:120~138.
[82]李彦锋.基于图像处理的植物叶片几何形态数据获取方法的研究,[硕士学位论文].哈尔滨:东北农业大学,2006.
[83]张斌.基于特征的回转体零件特征库的创建.北京:机械工业出版社,2006,35(5):148~150.
[84]李世国. Pro/TOOLKIT程序设计.机械制造与自动化,2003:313~380.