机械工程图图形检索技术研究
|
摘要
将类似产品的成功设计方法应用在新产品设计中，能明显提高设计质量，加快产品开发速度，因此，在新产品的开发中，工程技术人员经常要查询已有的相关设计资料，通过这种方法来借鉴以往的设计方法和经验。工程图包涵了产品设计、加工、安装的详细信息，是制造行业在生产中使用的重要技术文件，如何才能有效地从大量的工程图中检索到类似产品的图纸资料，从而为新设计提供参考是一个具有实际意义的问题。
当前，计算机技术在产品设计中被广泛使用，尽管设计图纸的生成和保存方式都发生了显著的变化，但工程图的检索问题仍然没有被很好地解决。在基于内容的视觉信息检索技术和计算机数据管理技术研究中取得的研究成果为解决这个问题提供了有益的借鉴，使得快速查找所需的设计图纸成为可能。
本论文针对机械工程图的特点，利用已有的研究成果，对机械工程图检索中的有关问题进行了探索，提出了解决工程图图形检索问题的方法。
图形信息是工程图中的主要内容，本课题重点研究了机械工程图中的图形信息，提出了工程图检索系统的总体结构框图，对系统各模块进行了分析，设计了各模块的功能，编制程序实现了关键功能。
为了保证轮廓图形的完整，针对一些会破坏图形完整性的表示方法，分析了视图的特征，提出了对应的轮廓图形恢复方法。研究了CAD工程图形文件的数据存储方式和ObjectARX函数库的使用方法，使用ObjectARX接口函数库编制程序，实现了从AutoCAD图形文件中提取所需轮廓图形信息的功能。
通过研究机械工程图图形的结构特点，将复杂图形分解成为复合图形、链式图形和简单图形的分层结构形式。提出了复合图形、链式图形的提取算法，借鉴链式图形分解为简单图形的实现方法，将描述机件轮廓的复杂图形最终表示为三种图形在二维空间的组合形式。
论文最后研究了如何有效地记录和组织图形描述信息，给出了图形的相似程度的评判方式及其数学表达形式，并提出了通过图形描述信息匹配来实现图形检索的策略。
上述研究工作发展了工程图计算机检索技术，提供了机械工程图图形检索的实现方法和步骤，为工程图图形检索技术向实际应用发展提供了参考。
The design quality can be improved and the development speed can be accelerated obviously by applying successful design methods of similar products to the development of new product. Engineering technicians like to query design data of relevant products in the process of developing a new product, by this means, they can refer previous designing methods and experiences. Engineering drawing, the main carrier of design, producing and installation information, is one of the important technical document that used wildly in manufacturing. It is a practical problem that how the designers to retrieve design drawing of similar products from a large number of drawing files and obtain useful reference for new design.
Though computer technology have be wildly used in product design currently, the information generation manner and storage form of engineering drawing are distinctly different to those of before, the problem how to retrieve required engineering drawing from drawing library still has not been solved perfectly. The technology, which is researched in context-based visual information retrieval and computer data management, gives some reference to this problem. Taking advantage of the researching production in those two fields, it is possible to achieve retrieving queried engineering drawing from many drawing files quickly.
Based on analyzing the particularity of mechanical engineering drawing and previous research results, this thesis researches the correlative aspects of mechanical engineering drawing retrieval and presents corresponding resolvents.
By analyzing the structure and the context of engineering drawing, this thesis ascertains the research emphasis of drawing shape retrieval is shape information. The fundamental framework of engineering drawing shapes retrieval system is presented and the ways to implement the functions of main modules are advanced as well.
In order to assure the contour is complete, for some representations that may destroy the integrity of contour view, the characteristic of views are analyzed and the ways to resume incomplete contours are presented. Based on describing the data storage structure of CAD engineering drawing, an application that can extract required contour shapes from CAD drawing under AutoCAD environment was developed by ObjectARX interface function library.
Based on analyzing the characteristic of the shape structure of engineering drawing, this thesis presents the methods to describe the complex contour by hierarchical structure that consist of compound shape, chain shape and simple shape. The extracting methods of three shapes put forward in above are presented. By this way, the context of complex contour shape is described by spatial combination of those three shapes.
At the end of the thesis, the methods to record and organize the describing information of shapes are researched. The criterion and the math expression of similarity metric for estimating the comparability among shape are given. The strategy to retrieve similar shape by
matching description information is presented as well.
This research develops the retrieval technology of engineering drawing and presents the implement methods and process of mechanical drawing retrieval which give reference to the practical application of engineering drawing retrieval.
当前，计算机技术在产品设计中被广泛使用，尽管设计图纸的生成和保存方式都发生了显著的变化，但工程图的检索问题仍然没有被很好地解决。在基于内容的视觉信息检索技术和计算机数据管理技术研究中取得的研究成果为解决这个问题提供了有益的借鉴，使得快速查找所需的设计图纸成为可能。
本论文针对机械工程图的特点，利用已有的研究成果，对机械工程图检索中的有关问题进行了探索，提出了解决工程图图形检索问题的方法。
图形信息是工程图中的主要内容，本课题重点研究了机械工程图中的图形信息，提出了工程图检索系统的总体结构框图，对系统各模块进行了分析，设计了各模块的功能，编制程序实现了关键功能。
为了保证轮廓图形的完整，针对一些会破坏图形完整性的表示方法，分析了视图的特征，提出了对应的轮廓图形恢复方法。研究了CAD工程图形文件的数据存储方式和ObjectARX函数库的使用方法，使用ObjectARX接口函数库编制程序，实现了从AutoCAD图形文件中提取所需轮廓图形信息的功能。
通过研究机械工程图图形的结构特点，将复杂图形分解成为复合图形、链式图形和简单图形的分层结构形式。提出了复合图形、链式图形的提取算法，借鉴链式图形分解为简单图形的实现方法，将描述机件轮廓的复杂图形最终表示为三种图形在二维空间的组合形式。
论文最后研究了如何有效地记录和组织图形描述信息，给出了图形的相似程度的评判方式及其数学表达形式，并提出了通过图形描述信息匹配来实现图形检索的策略。
上述研究工作发展了工程图计算机检索技术，提供了机械工程图图形检索的实现方法和步骤，为工程图图形检索技术向实际应用发展提供了参考。
The design quality can be improved and the development speed can be accelerated obviously by applying successful design methods of similar products to the development of new product. Engineering technicians like to query design data of relevant products in the process of developing a new product, by this means, they can refer previous designing methods and experiences. Engineering drawing, the main carrier of design, producing and installation information, is one of the important technical document that used wildly in manufacturing. It is a practical problem that how the designers to retrieve design drawing of similar products from a large number of drawing files and obtain useful reference for new design.
Though computer technology have be wildly used in product design currently, the information generation manner and storage form of engineering drawing are distinctly different to those of before, the problem how to retrieve required engineering drawing from drawing library still has not been solved perfectly. The technology, which is researched in context-based visual information retrieval and computer data management, gives some reference to this problem. Taking advantage of the researching production in those two fields, it is possible to achieve retrieving queried engineering drawing from many drawing files quickly.
Based on analyzing the particularity of mechanical engineering drawing and previous research results, this thesis researches the correlative aspects of mechanical engineering drawing retrieval and presents corresponding resolvents.
By analyzing the structure and the context of engineering drawing, this thesis ascertains the research emphasis of drawing shape retrieval is shape information. The fundamental framework of engineering drawing shapes retrieval system is presented and the ways to implement the functions of main modules are advanced as well.
In order to assure the contour is complete, for some representations that may destroy the integrity of contour view, the characteristic of views are analyzed and the ways to resume incomplete contours are presented. Based on describing the data storage structure of CAD engineering drawing, an application that can extract required contour shapes from CAD drawing under AutoCAD environment was developed by ObjectARX interface function library.
Based on analyzing the characteristic of the shape structure of engineering drawing, this thesis presents the methods to describe the complex contour by hierarchical structure that consist of compound shape, chain shape and simple shape. The extracting methods of three shapes put forward in above are presented. By this way, the context of complex contour shape is described by spatial combination of those three shapes.
At the end of the thesis, the methods to record and organize the describing information of shapes are researched. The criterion and the math expression of similarity metric for estimating the comparability among shape are given. The strategy to retrieve similar shape by
matching description information is presented as well.
This research develops the retrieval technology of engineering drawing and presents the implement methods and process of mechanical drawing retrieval which give reference to the practical application of engineering drawing retrieval.
引文
[1] 王玉，邢渊，阮雪榆．机械产品设计重用策略研究．机械工程学报，2002，38(5)：145～148
[2] 阮雪榆．工程设计重用理论及其关键技术研究(博士学位论文)．上海交通大学，2001
[3] 丁宁明．工程图学学科内涵及分支探讨．工程图学学报，1998，19(1)：84～89
[4] 王成良．图形关联信息的数据库查询及其应用．重庆大学学报(自然科学版)，2002，25(1)：57～60
[5] 吕志军，杨建国，魏虹玢．GTCAD中特征建模的应用研究．机械制造，2003，41(11)：21～23
[6] 孙战里，尹延国，朱元吉．成组技术思想在塑料异型材挤出模功能块结构设计中的应用，化学建材，2002，18(4)：3～5
[7] Tanveer Syeda-Mahmood. Locating indexing structures in engineering drawing databases using location hashing. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1999, (1):49～55
[8] Stefan Müller, Gerhard Rigoll. Searching an engineering drawing database for user-specified shapes. Document Analysis and Recognition, Proceedings of the Fifth International Conference, 1999, (1): 697～700
[9] Stefan Berchtold, Hans-Peter Kriegel. S3: Similarity search in CAD database systems. Management of Data, Proceedings of the 1997 ACM SIGMOD International Conference. 1997, 26 (2): 564～567
[10] Manuel J. Fonseca, Joaquim A. Jorge. Towards content-based retrieval of technical drawings through high-dimensional indexing. Computers and Graphics, 2003, 27(1): 61～69
[11] Jong H. Park, Bong S. Um, A new approach to similarity retrieval of 2-D graphic objects based on dominant shapes. Pattern Recognition Letters, 1999, 20(6): 591～616.
[12] Mohamed E1-Mehalawi, R. Allen Miller. A database system of mechanical components based on geometric and topological similarity. Part Ⅰ: representation. Computer-Aided Design, 2003, 35(1): 83-105
[13] W.C. Regli, V.A. Cicirello, Managing digital libraries for computer-aided design. Computer-Aided Design, 2000, 32 (2): 119～132
[14] 宋海玉．简单CAD图纸检索的研究与实现(硕士学位论文)．吉林大学，2003
[15] 马建，滕弘飞，刘德权．基于样图的排样及其样图检索方法．软件学报，2000，11(12)：1685～1691
[16] 翁建厂，庄越挺，潘云鹤．一种三维部件形状检索方法．机械工程学报，2003，39(10)：117～122
[17] 章毓晋．基于内容的视觉信息检索．北京：科学出版社．2003
[18] 李蓉，张树生．扫描工程图二维重建中基于图表达的结构约束校正．计算机工程与科学，2001，23(6)，22～24
[19] Flickner etal.Query by image and video content:the OBIC system.IEEE Computer,1995,(9):23～31.
[20] 徐杰，施鹏飞．基于内容的图象检索技术．中国图象图形学报：A辑，2003，8(9)：977～983
[21] 任金吕，赵朵椿，冯大淦．用形状相似性进行基于内容的图象检索．中国体视学与图像分析，2000，5(1)：44～48
[22] 黄元元，郭丽，杨静宇．利用形状与空间位置特征检索二值商标图象．中国图象图形学报：A辑，2002，7(11)：1187～1191
[23] 张文景，许晓鸣，苏键锋．基于Hausdorff距离的2D形状匹配改进算法．中国图象图形学报，2000，05A(2)：106～109
[24] 张量．一种基于对象空间关系的图像检索模型．计算机工程，2002，28(4)：175～177，182
[25] Chang S.K., Donato N., McCormick B.H., A Relational Database System For Pictures, IEEE Workshop on Pictures Data Description and Management, 1977, (1): 231～236
[26] 潘健，杨天．AutoCAD 2002机械图形绘制精解．北京：人民邮电出版社，2001
[27] 邵俊吕，李旭东编．AutoCAD ObJectARX 2000开发技术指南．北京：电子工业出版社，2000
[28] AutoDesk. ObjectARX forAutoCAD 2002 Online Help. Autodesk, Inc., 2001
[29] 利克智．C++程序设计．西安：西安交通大学出版社，1995
[30] 国家标准化管理委员会．机械工程CAD制图规则(国家标准)．北京：中国标准出版社，1998
[31] 大连理工大学工程画教研室．机械制图．北京：高等教育出版社，2000
[32] 赵敬千．工程图纸扫描输入与识别系统的研究与实现[硕士学位论文]．西安交通大学，2000
[33] Y.B. Bai, X.W. Xu. Object Boundary Encoding—a new vectorisation algorithm for engineering drawings. Computers in Industry, 2001, 46(1): 65～74.
[34] Yuan Chen, Noshir A. Langrana, Atish K. Das. Perfecting Vectorized Mechanical Drawings. Computer Vision and Image Understanding, 1996, 63 (2): 273～286,
[35] 郁晓红，李伟青．工程图纸中虚线的提取算法．计算机应用与软件，1999，16(3)：16～19，35
[36] Canny J F. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1985, 18(6): 679～698
[37] 贾云德．机器视觉．北京：科技出版社，2000
[38] Maria P. Optimal edge detectors for ramp edge. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1991, 13(5): 483～491
[39] 苏丰，宋继强，陈冀兵，蔡世杰．工程图纸矢量化软件VH-Vector用户接口的设计与实现．计算机学报，2001，24(9)：930～937
[40] Softelec. VPHybridCAD V7-Overview. http://www.softelec.com/
[41] 陆国栋，张树有等．工程图样数字化转换与智能理解，北京：机械出版社，2001，5
[42] 何有钧，邹慧君．基于图形分解思想的复杂零件的参数化设计．机械科学与技术(西安)，2000，19(2)：201～202
[43] 史奇，钟万勰．基于人工智能原理的二维图形分解．计算机工程，1990，(2)：41～46，70
[44] 张树有．侧点法自动跟踪获取图形轮廓信息．浙江大学学报：自然科学版，1996，30(4)：403～407
[45] 张建勋，何玉林．零件二维视图轮廓信息和封闭图形信息的自动提取．计算机工程与应用，2000，36(10)：38～40
[46] 张淮声，张佑生，方贤勇．基于矢量积的二维封闭图形轮廓信息提取方法．计算机工程与应用，2002，38(8)：93～94，151
[47] 杨岳，陈峰，罗意平．AutoCAD环境下二维图形轮廓的自动识别与生成．长沙铁道学院学报，2003，21(3)：67～71
[48] 刘云华，钟毅芳，秦红斌．基于模式分解和符号匹配的特征识别方法研究．机械科学与技术(西安)．2000，19(5)：845～848
[49] Adam Drozdek．数据结构与算法——C++版(第2版)(影印版)．北京：清华大学出版社
[50] Jia Wang, Wen jann Yang, Raj Acharya. Efficient access to and retrieval from a shape image database. In Proceedings of Content-Based Access of Image and Video Libraries, Santa Barbara, CA, 1998, (1): 63～67.
[51] 孙季丰，方少元，几何型哈什法和图像的快速合成．华南理工大学学报(自然科学版)，1998，26(2)：75～80.
[52] R. Jain. Similarity measures for image databases. In Proceedings of IEEE International Conference on Fuzzy System, 1995, (3): 1247～1254
[53] 梅承力，杨景涛，周源华．基于空间数据结构的图像检索．上海交通大学学报，2002，36(6)：796-798，811
[54] 马志锋，郑晓妹．基于相似Rough集的模糊检索策略．计算机工程与应用，2000，36(2)：123～125
[55] 伯晓晨，黄爱民．图象内容检索中的相似性度量和分层索引．计算机工程与应用，1999，35(1)：58～60
[56] 韩水华，陈传波，陆正鼎．复杂相似性图像检索系统体系结构．华中科技大学学报：自然科学版，2001，29(3)：36～38，42
[2] 阮雪榆．工程设计重用理论及其关键技术研究(博士学位论文)．上海交通大学，2001
[3] 丁宁明．工程图学学科内涵及分支探讨．工程图学学报，1998，19(1)：84～89
[4] 王成良．图形关联信息的数据库查询及其应用．重庆大学学报(自然科学版)，2002，25(1)：57～60
[5] 吕志军，杨建国，魏虹玢．GTCAD中特征建模的应用研究．机械制造，2003，41(11)：21～23
[6] 孙战里，尹延国，朱元吉．成组技术思想在塑料异型材挤出模功能块结构设计中的应用，化学建材，2002，18(4)：3～5
[7] Tanveer Syeda-Mahmood. Locating indexing structures in engineering drawing databases using location hashing. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1999, (1):49～55
[8] Stefan Müller, Gerhard Rigoll. Searching an engineering drawing database for user-specified shapes. Document Analysis and Recognition, Proceedings of the Fifth International Conference, 1999, (1): 697～700
[9] Stefan Berchtold, Hans-Peter Kriegel. S3: Similarity search in CAD database systems. Management of Data, Proceedings of the 1997 ACM SIGMOD International Conference. 1997, 26 (2): 564～567
[10] Manuel J. Fonseca, Joaquim A. Jorge. Towards content-based retrieval of technical drawings through high-dimensional indexing. Computers and Graphics, 2003, 27(1): 61～69
[11] Jong H. Park, Bong S. Um, A new approach to similarity retrieval of 2-D graphic objects based on dominant shapes. Pattern Recognition Letters, 1999, 20(6): 591～616.
[12] Mohamed E1-Mehalawi, R. Allen Miller. A database system of mechanical components based on geometric and topological similarity. Part Ⅰ: representation. Computer-Aided Design, 2003, 35(1): 83-105
[13] W.C. Regli, V.A. Cicirello, Managing digital libraries for computer-aided design. Computer-Aided Design, 2000, 32 (2): 119～132
[14] 宋海玉．简单CAD图纸检索的研究与实现(硕士学位论文)．吉林大学，2003
[15] 马建，滕弘飞，刘德权．基于样图的排样及其样图检索方法．软件学报，2000，11(12)：1685～1691
[16] 翁建厂，庄越挺，潘云鹤．一种三维部件形状检索方法．机械工程学报，2003，39(10)：117～122
[17] 章毓晋．基于内容的视觉信息检索．北京：科学出版社．2003
[18] 李蓉，张树生．扫描工程图二维重建中基于图表达的结构约束校正．计算机工程与科学，2001，23(6)，22～24
[19] Flickner etal.Query by image and video content:the OBIC system.IEEE Computer,1995,(9):23～31.
[20] 徐杰，施鹏飞．基于内容的图象检索技术．中国图象图形学报：A辑，2003，8(9)：977～983
[21] 任金吕，赵朵椿，冯大淦．用形状相似性进行基于内容的图象检索．中国体视学与图像分析，2000，5(1)：44～48
[22] 黄元元，郭丽，杨静宇．利用形状与空间位置特征检索二值商标图象．中国图象图形学报：A辑，2002，7(11)：1187～1191
[23] 张文景，许晓鸣，苏键锋．基于Hausdorff距离的2D形状匹配改进算法．中国图象图形学报，2000，05A(2)：106～109
[24] 张量．一种基于对象空间关系的图像检索模型．计算机工程，2002，28(4)：175～177，182
[25] Chang S.K., Donato N., McCormick B.H., A Relational Database System For Pictures, IEEE Workshop on Pictures Data Description and Management, 1977, (1): 231～236
[26] 潘健，杨天．AutoCAD 2002机械图形绘制精解．北京：人民邮电出版社，2001
[27] 邵俊吕，李旭东编．AutoCAD ObJectARX 2000开发技术指南．北京：电子工业出版社，2000
[28] AutoDesk. ObjectARX forAutoCAD 2002 Online Help. Autodesk, Inc., 2001
[29] 利克智．C++程序设计．西安：西安交通大学出版社，1995
[30] 国家标准化管理委员会．机械工程CAD制图规则(国家标准)．北京：中国标准出版社，1998
[31] 大连理工大学工程画教研室．机械制图．北京：高等教育出版社，2000
[32] 赵敬千．工程图纸扫描输入与识别系统的研究与实现[硕士学位论文]．西安交通大学，2000
[33] Y.B. Bai, X.W. Xu. Object Boundary Encoding—a new vectorisation algorithm for engineering drawings. Computers in Industry, 2001, 46(1): 65～74.
[34] Yuan Chen, Noshir A. Langrana, Atish K. Das. Perfecting Vectorized Mechanical Drawings. Computer Vision and Image Understanding, 1996, 63 (2): 273～286,
[35] 郁晓红，李伟青．工程图纸中虚线的提取算法．计算机应用与软件，1999，16(3)：16～19，35
[36] Canny J F. A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1985, 18(6): 679～698
[37] 贾云德．机器视觉．北京：科技出版社，2000
[38] Maria P. Optimal edge detectors for ramp edge. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1991, 13(5): 483～491
[39] 苏丰，宋继强，陈冀兵，蔡世杰．工程图纸矢量化软件VH-Vector用户接口的设计与实现．计算机学报，2001，24(9)：930～937
[40] Softelec. VPHybridCAD V7-Overview. http://www.softelec.com/
[41] 陆国栋，张树有等．工程图样数字化转换与智能理解，北京：机械出版社，2001，5
[42] 何有钧，邹慧君．基于图形分解思想的复杂零件的参数化设计．机械科学与技术(西安)，2000，19(2)：201～202
[43] 史奇，钟万勰．基于人工智能原理的二维图形分解．计算机工程，1990，(2)：41～46，70
[44] 张树有．侧点法自动跟踪获取图形轮廓信息．浙江大学学报：自然科学版，1996，30(4)：403～407
[45] 张建勋，何玉林．零件二维视图轮廓信息和封闭图形信息的自动提取．计算机工程与应用，2000，36(10)：38～40
[46] 张淮声，张佑生，方贤勇．基于矢量积的二维封闭图形轮廓信息提取方法．计算机工程与应用，2002，38(8)：93～94，151
[47] 杨岳，陈峰，罗意平．AutoCAD环境下二维图形轮廓的自动识别与生成．长沙铁道学院学报，2003，21(3)：67～71
[48] 刘云华，钟毅芳，秦红斌．基于模式分解和符号匹配的特征识别方法研究．机械科学与技术(西安)．2000，19(5)：845～848
[49] Adam Drozdek．数据结构与算法——C++版(第2版)(影印版)．北京：清华大学出版社
[50] Jia Wang, Wen jann Yang, Raj Acharya. Efficient access to and retrieval from a shape image database. In Proceedings of Content-Based Access of Image and Video Libraries, Santa Barbara, CA, 1998, (1): 63～67.
[51] 孙季丰，方少元，几何型哈什法和图像的快速合成．华南理工大学学报(自然科学版)，1998，26(2)：75～80.
[52] R. Jain. Similarity measures for image databases. In Proceedings of IEEE International Conference on Fuzzy System, 1995, (3): 1247～1254
[53] 梅承力，杨景涛，周源华．基于空间数据结构的图像检索．上海交通大学学报，2002，36(6)：796-798，811
[54] 马志锋，郑晓妹．基于相似Rough集的模糊检索策略．计算机工程与应用，2000，36(2)：123～125
[55] 伯晓晨，黄爱民．图象内容检索中的相似性度量和分层索引．计算机工程与应用，1999，35(1)：58～60
[56] 韩水华，陈传波，陆正鼎．复杂相似性图像检索系统体系结构．华中科技大学学报：自然科学版，2001，29(3)：36～38，42