首页> 外文学位 >Interpretation of engineering drawings of polyhedral and nonpolyhedral objects from orthographic projections.
【24h】

Interpretation of engineering drawings of polyhedral and nonpolyhedral objects from orthographic projections.

机译:从正投影中解释多面体和非多面体对象的工程图。

获取原文
获取原文并翻译 | 示例

摘要

A system is presented for the interpretation of multiple-view engineering drawings. This system takes a set of two-dimensional orthographic projections and reconstructs the three-dimensional shape of the object, using a boundary representation model. Both polyhedral objects and objects with curved surfaces (spherical or cylindrical) can be interpreted. For its input, the system requires a set of two-dimensional lines as a description of the drawing to be interpreted. This type of description can be obtained from a system previously developed by Kasturi et al., which takes a paper-based drawing as input, recognizes 2D objects, and produces a list of lines contained in that drawing.; The system can accommodate auxiliary views in addition to the standard orthogonal views, and the layout of the views need not be known a priori. A tree representation is developed for describing the relationships between the views. A basis matrix formalism for relating the 2D view-based coordinates to the 3D object-based coordinates is also presented.; A general method is introduced which, unlike previous methods, avoids any assumptions about the nature of any nonplanar surfaces until a later stage of the process when more reliable decisions can be made. A new technique based on the Dempster-Shafer theory of evidence is introduced to guide the construction of the 3D model. This technique accommodates the additional complexity that results from deferring decisions about the object structure and also allows drawings with missing or omitted lines to be reasonably interpreted when information about the features they represent is available from other views. The system can also handle drawings with moderate errors in the alignment of the views and in the location of the input lines.
机译:提出了一种用于解释多视图工程图的系统。该系统采用一组二维正射投影,并使用边界表示模型重建对象的三维形状。多面体和曲面(球面或圆柱体)都可以解释。对于其输入,系统需要一组二维线作为要解释的附图的描述。这种类型的描述可以从Kasturi等人先前开发的系统中获得,该系统将基于纸的图形作为输入,识别2D对象,并生成该图形中包含的线的列表。除了标准正交视图之外,该系统还可以容纳辅助视图,并且视图的布局无需事先知道。开发了一种树形表示形式,用于描述视图之间的关系。还提出了用于将基于2D视图的坐标与基于3D对象的坐标相关联的基本矩阵形式。引入了一种通用方法,该方法与以前的方法不同,它可以避免对任何非平面表面的性质做出任何假设,直到可以做出更可靠决策的过程的后期为止。引入了基于Dempster-Shafer证据理论的新技术来指导3D模型的构建。此技术可适应因推迟对对象结构的决策而导致的额外复杂性,并且还可以在其他视图中获得有关其表示的特征的信息时,合理地解释带有缺失或遗漏线条的图形。该系统还可以处理视图对齐和输入线位置中有中等误差的图形。

著录项

  • 作者

    Lysak, Daniel Bohdan, Jr.;

  • 作者单位

    The Pennsylvania State University.;

  • 授予单位 The Pennsylvania State University.;
  • 学科 Computer Science.; Mathematics.
  • 学位 Ph.D.
  • 年度 1991
  • 页码 183 p.
  • 总页数 183
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 自动化技术、计算机技术;数学;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号