基于神经网络边缘提取的工业CT图像与CAD模型的比对检测

曾理; 何洪举; 张志波

doi:10.3788/OPE.20111910.2533

您当前的位置：

首页 >

文章列表页 >

基于神经网络边缘提取的工业CT图像与CAD模型的比对检测

信息科学 | 更新时间：2020-08-12

- 基于神经网络边缘提取的工业CT图像与CAD模型的比对检测
- Comparison inspection between ICT images & CAD model based on edge extracting by neural networks
- 光学精密工程 2011年19卷第10期页码：2533-2540
- 作者机构：
  
  1. 重庆大学数学与统计学院, 重庆 401331
  2. 重庆大学光电技术及系统教育部重点实验室 ICT研究中心, 重庆 400030
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.60972104)();重庆市自然科学基金资助项目(No.2010BB4222)()
- DOI：10.3788/OPE.20111910.2533
  中图分类号： TP183;TP391
- 收稿日期：2011-01-18，
  
  修回日期：2011-02-27，
  
  网络出版日期：2011-10-27，
  
  纸质出版日期：2011-10-25
- 稿件说明：
移动端阅览
曾理, 何洪举, 张志波. 基于神经网络边缘提取的工业CT图像与CAD模型的比对检测[J]. 光学精密工程, 2011,19(10): 2533-2540

ZENG Li, HE Hong-ju, ZHANG Zhi-bo . Comparison inspection between ICT images & CAD model based on edge extracting by neural networks[J]. Editorial Office of Optics and Precision Engineering, 2011,19(10): 2533-2540
曾理, 何洪举, 张志波. 基于神经网络边缘提取的工业CT图像与CAD模型的比对检测[J]. 光学精密工程, 2011,19(10): 2533-2540 DOI： 10.3788/OPE.20111910.2533.

ZENG Li, HE Hong-ju, ZHANG Zhi-bo . Comparison inspection between ICT images & CAD model based on edge extracting by neural networks[J]. Editorial Office of Optics and Precision Engineering, 2011,19(10): 2533-2540 DOI： 10.3788/OPE.20111910.2533.

摘要

研究了一种对工业计算机断层(CT)图像与计算机辅助设计(CAD)模型进行比对检测

分析工件制造误差的方法。首先

用模板自适应细胞神经网络提取工业CT图像边缘

并进行三方向CT切片边缘数据融合处理以获得完整的三维边缘面。然后

先结合主成分分析和最小包围盒的思想对CT边缘面数据与工件的CAD模型实现粗配准

再用奇异值分解-迭代最近点算法对其进行精配准

其中最近点对的求取用

k-d

树进行加速

从而实现对工件制造误差的分析。实验结果显示

文中的方法能够实现工件的比对检测

自动化程度高、能直观显示误差分布且精度高

表明通过改进工件CT图像与CAD模型的比对检测方法

可将工业CT技术用于制造工艺分析与改进中。

Abstract

A method to analyze the manufacture error of a workpiece based on the comparison inspection between Industrial Computed Tomography (ICT) images and Computer Aided Design (CAD) model was discussed. Firstly

the edged surfaces of ICT images were extracted by the Cellular Neural Network (CNN) with adaptive templates and the data were fused in three directions to obtaine the complete 3D edge surfaces. Then

the Principal Component Analysis (PCA) with the method of minimum bounding box were combined to perform a rough registration

and Singular Value Decomposition and Iterative Closest Point (SVD-ICP) algorithm were used to realize the refined registration for the edged surface data and the CAD model. In experiment

the

k-d

tree was used to improve the calculation speed of searching for the closest point. The experimental results validate that the comparison inspection method is automatic

visualized and high-accuracy. By the improved comparison inspection method for ICT images and CAD model

the ICT technology can be used to analyze and improve the manufacturing process.

关键词

Keywords

references

SHAMMAA M H, SUZUKI H, MICHIKAWA T. Registration of CAD mesh models with CT volumetric model of assembly of machine parts[J]. Visual Compute, 2007(23):965-974.[2] 刘晶,张定华,毛海鹏. 基于复形法配准工业产品CT切片模型和CAD模型[J]. 计算机应用研究,2004(7):248-252. LIU J, ZHANG D H, MAO H P. Registration of industrial products CT slice model and CAD model based on the method of complex [J]. Application Research of Computers, 2004(7):248-252.(in Chinese)[3] 胡栋材,赵歆波,张定华,等. 自适应遗传算法配准三维数字样品模型及其CAD模型[J]. CT理论与应用研究.2008, 17(2):8-14. HU D C, ZHAO X B, ZHANG D H, et al.. Registration of three dimension digital model and CAD model based on the method of adaptive genetic algorithm[J]. CT Theory and Applications, 2008, 17(2):8-14.(in Chinese)[4] 刘晶,张定华,毛海鹏,等. 基于SVD-ICP算法配准CT切片重构模型与CAD模型[J]. 计算机工程与应用,2004,24:195-204. LIU J, ZHANG D H, MAO H P, et al.. Registration of the reconstructed CT slice model and CAD model based on the SVD-ICP algorithm[J]. Computer Engineering and Applications, 2004,14:195-204. (in Chinese)[5] 邹晓兵,曾理. 锥束螺旋CT半覆盖扫描重建[J]. 光学精密工程, 2010,18(2):434-442. ZOU X B, ZENG L. Half-cover scanning and reconstructing for helical cone-beam CT [J]. Opt. Precision Eng., 2010, 18(2):434-442. (in Chinese)[6] 曾理,郭海燕,马睿. 基于小波定位及Facet模型的三维工业CT图像边缘检测[J]. 光学精密工程,2010,18(2):443-450. ZENG L, GUO H Y, MA R. Edge detection of industrial CT 3D image based on wavelet locating and Facet model [J]. Opt. Precision Eng., 2010,18(2):443-450.(in Chinese)[7] 曾理,何洪举,刘长江. 基于神经网络边缘提取的工业断层成像图像拟合[J]. 计算机集成制造系统,2010,16(1):24-29. ZENG L, HE H J, LIU CH J. Fitting of industrial computer tomography images base on edge extracting by neural networks[J]. Computer Integrated Manufacturing Systems, 2010,16(1):24-29.(in Chinese)[8] CHUA L O, YANG L. Cellular neural networks: Theory [J]. IEEE Trans. Circuits Syst, 1988,35(10):1257-1272.[9] CHUA L O, YANG L. Cellular neural network s: Application [J]. IEEE Trans. Circuits Syst, 1988, 35(10):1273-1290.[10] 黄蕾,刘文波. 基于CNN的灰度图像边缘提取算法中模板参数的研究[J]. 计算机与现代化, 2006(6):69-74. HUANG L, LIU W B. Research on template parameters of edge detection of gray-scale images based on cellular neural networks[J]. Computer and Modernization, 2006(6):69-74.(in Chinese)[11] 王明敏,周兆敏,张侃瑜,等. 基于模板自适应细胞神经网络的图像处理及识别[J]. 上海大学学报(自然科学版), 1998,4(2):175-179. WANG M M, ZHOU ZH M, ZHANG K Y, et al.. Image processing and image recognition based on the cellular neural network with the adaptive templates[J]. Journal of Shanghai University(Natural Science), 1998, 4(2):175-179.(in Chinese)[12] 刘长江,曾理. 工业CT三维体数据边缘面提取[J]. 计算机工程与应用, 2009, 45(18): 219-221. LIU CH J, ZENG L. Edged surface extraction of 3D volume data from industrial CT[J]. Computer Engineering and Applications, 2009, 45(18):219-221.(in Chinese)[13] 刘晶. 基于CAD模型的配准技术综述[J]. 机床与液压, 2007, 35(9):231-232. LIU J. Review of registration based on CAD model[J]. Machine Tool & Hydraulics, 2007, 35(9):231-232.(in Chinese)[14] 张开兴,张树生,白晓亮. 基于最小包围盒的三维模型的配准技术[J]. 机床与液压, 2008, 36(4):5-7. ZHANG K X, ZHANG SH SH, BAI X L. Technique of 3D models for registration based on minimal bounding box [J]. Machine Tool & Hydraulics, 2008,36(4):5-7.(in Chinese)[15] 许法强,万晓霞,朱元泓. 基于旋转主成分分析的颜色组分预测[J]. 光学精密工程, 2008, 16(3):518-523. XU F Q, WAN X X, ZHU Y H. Color component prediction based on rotated principal component analysis[J]. Opt. Precision Eng., 2008, 16(3):518-523.(in Chinese)[16] BESL P J, MCKAY N D. A method for registration of 3-D shapes[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1992,14(2):239-256.[17] UMEYAMA S. Least-squares estimation of transformation parameters between two point patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1991,13(4):376-380.[18] 史宝全,梁晋,刘青,等. 基于约束搜索球的点云数据与CAD模型精确比对检测[J]. 计算机集成制造系统, 2010, 16(5):929-934. SHI B Q, LIANG J, LIU Q, et al.. Precision inspection of point cloud & CAD model based on constraint search sphere[J]. Computer Integrated Manufacturing Systems, 2010, 16(5):929-934.(in Chinese)[19] 中国科学院自动化研究所医学图像研究室. 3DMed . (2009-12-31).http://www.3dmed.net. Medical Image Processing Group, Institute of Automation, Chinese Academy of Sciences. 3DMed . ( 2009-12-31). http://www.3dmed.net. (in Chinese)[20] 段黎明,刘元宝,吴志芳,等. 基于工业计算机断层成像技术的三维CAD 模型重构方法[J]. 计算机集成制造系统, 2009,15(3): 479-486. DUAN L M, LIU Y B, WU Z F, et al.. Method of reconstructing 3D CAD model based on industrial computed tomography [J]. Computer Integrated Manufacturing Systems, 2009, 15(3):479-486.(in Chinese)

浏览量

399

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于神经网络的全斯托克斯光电探测器

基于圆卷积神经网络的粘连导电粒子检测

基于自适应局部邻域条件下的点云匹配

面向领域自适应的部分最优传输高光谱图像分类

基于多特征参数融合的文物碎片自动匹配方法