逆向工程中基于属性邻接图的加工特征识别

• 论文 •

逆向工程中基于属性邻接图的加工特征识别

刘雪梅，张树生, 崔卫卫，　宋加玉

1.西北工业大学现代设计与集成制造技术教育部重点实验室，陕西西安710072；2.华北水利水电学院信息工程系, 河南郑州450011

出版日期:2008-06-15 发布日期:2008-06-25

Machined features recognition based on attributed adjacency graph in reverse engineering

LIU Xue-mei, ZHANG Shu-sheng, CUI Wei-wei, SONG Jia-yu

1.Ministry of Education Key Lab of Contemporary Design & Integrated Manufacturing Technology, Northwestern Polytechnical University, Xi’an 710072, China; 2. Department of Information Engineering, North China Institute of Water Conservancy and Hydroelectric Power, Zhengzhou 450011, China

Online:2008-06-15 Published:2008-06-25

摘要/Abstract

摘要： 为了提供逆向工程系统和CAx系统集成的直接通道，提出了针对点云数据建立属性邻接图的算法，并利用属性邻接图实现加工特征自动识别。首先利用模糊c\|均值聚类方法对点云进行分区，将分区后的面片进行曲面类型判别和几何参数提取。设计了建立面片之间邻接关系和判断相邻面片之间凸凹性的算法，从而建立零件的属性邻接图；对零件的属性邻接图进行分解，遍历分解后的子图形成特征子图，将特征子图与特征规则相匹配，从而实现加工特征自动识别。实验结果证明了该算法的有效性。

关键词: 加工特征, 特征识别, 点云, 逆向工程

Abstract: To offer a direct path for integration of reverse engineering and CAx system, machined features recognition based point cloud data was put forward. The algorithm of constructing Attributed Adjacency Graph (AAG) based point cloud data was presented, and then AAG was used to recognize machined features. Firstly, fuzzy c-means cluster method was used to segment point cloud. Surface types of patches were distinguished and their geometric parameters were extracted. The algorithm of constructing adjacency relationships and the method of judging salient or nook among adjacency patches were designed, thus AAG of part was set up. The AAG was decomposed and traversed to form feature sub-graphs. The sub-graphs were matched to feature rules to fulfill automatic recognition of machined feature. Experimental results showed that the algorithm was effective.

Key words: machined feature, feature recognition, point cloud, reverse engineering

中图分类号:

TP391

刘雪梅,，张树生, 崔卫卫，　宋加玉. 逆向工程中基于属性邻接图的加工特征识别[J]. .

LIU Xue-mei,, ZHANG Shu-sheng, CUI Wei-wei, SONG Jia-yu. Machined features recognition based on attributed adjacency graph in reverse engineering[J]. .

[1]	李学兵,胡天亮,刘忠强,马嵩华. 基于工件特征的曲面生成优化方法[J]. 计算机集成制造系统, 2021, 27(8): 2196-2203.
[2]	杨通,姚山,薛凯华. 用于增材制造的点云直接分层算法[J]. 计算机集成制造系统, 2021, 27(5): 1371-1381.
[3]	王耀,王春香,周国勇,纪康辉,刘流,郝林文. 快速成型技术中分层算法的研究综述[J]. 计算机集成制造系统, 2021, 27(3): 828-841.
[4]	祝鹏,余建波,郑小云,孙习武,王永松,陈长江. 混合机械加工过程的偏差网络建模与误差溯源[J]. 计算机集成制造系统, 2020, 26(第3): 664-675.
[5]	华顺刚,许林林,白茂东. 零件模型的铸件毛坯自动生成方法[J]. 计算机集成制造系统, 2018, 24(第11): 2827-2835.
[6]	王武礼,段黎明,王浩宇,白洋. 基于动态误差控制和PSO的三角网格模型简化优化方法[J]. 计算机集成制造系统, 2018, 24(第1): 63-71.
[7]	余飞祥,黄翔,李泷杲,李栋. 基于扫描线点云的飞机蒙皮边界特征提取[J]. 计算机集成制造系统, 2017, 23(第4期): 701-707.
[8]	刘金锋,吴健,周宏根,田桂中,刘晓军. 基于加工特征的工艺信息重用方法[J]. 计算机集成制造系统, 2017, 23(第4期): 791-798.
[9]	张龙,胡俊锋,熊国良. 基于MED和ICA的滚动轴承循环冲击故障特征增强[J]. 计算机集成制造系统, 2017, 23(第2期): 333-339.
[10]	林洪彬,付德敏,王银腾. 基于参数自适应各向异性高斯核的散乱点云保特征去噪[J]. 计算机集成制造系统, 2017, 23(第12): 2583-2592.
[11]	魏涛,张丹,左敦稳,徐锋,夏三星. 面向薄壁多腔类结构件加工特征识别方法[J]. 计算机集成制造系统, 2017, 23(第12): 2683-2691.
[12]	唐倩,黄耀,范秋垒,刘飞,裴林清. 基于点云与图像匹配的自动化喷涂系统坐标变换[J]. 计算机集成制造系统, 2017, 23(第11): 2392-2398.
[13]	高晓兵,万能. 基于NURBS方法的加工特征几何演变识别[J]. 计算机集成制造系统, 2016, 22(第8期): 1919-1928.
[14]	魏栋,张树有,刘晓健. 基于压缩控制顶点的B样条曲线逼近技术[J]. 计算机集成制造系统, 2016, 22(第6期): 1396-1402.
[15]	李春磊,莫蓉,常智勇,杨海成,张栋梁,向颖. 几何演变驱动的机加工艺知识表示与推送[J]. 计算机集成制造系统, 2016, 22(第6期): 1434-1446.