基于集成自适应欠采样的铝管表面缺陷检测方法研究

doi:10.3901/JME.2023.06.018

机械工程学报 ›› 2023, Vol. 59 ›› Issue (6): 18-31.doi: 10.3901/JME.2023.06.018

扫码分享

基于集成自适应欠采样的铝管表面缺陷检测方法研究

郎宁^1,2, 王德成¹, 程鹏¹

1. 中国机械科学研究总院中机生产力促进中心有限公司北京 100044;
2. 湖南大学汽车车身先进设计制造国家重点实验室长沙 410082

收稿日期:2022-08-24 修回日期:2022-11-18 出版日期:2023-03-20 发布日期:2023-06-03
通讯作者: 王德成(通信作者),男,1962年出生,博士,研究员,博士研究生导师。主要研究方向为可靠性分析和智能制造。E-mail:wangdc@cam.com.cn
作者简介:郎宁,男,1993年出生,博士研究生。主要研究方向为机器视觉。E-mail:ninglang@hnu.edu.cn
基金资助:
国家科技重大专项资助项目(2019ZX04021001)。

Research on Surface Defect Detection Method of Aluminum Tube Based on Ensemble Adaptive Undersampling

LANG Ning^1,2, WANG Decheng¹, CHENG Peng¹

1. China Productivity Centre for Machinery, China Academy of Machinery Science and Technology, Beijing 100044;
2. State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082

Received:2022-08-24 Revised:2022-11-18 Online:2023-03-20 Published:2023-06-03

摘要/Abstract

摘要： 铝管作为一种常见的传输零件，对其表面缺陷进行检测是保证其生产质量、运行安全的必要措施。基于机器视觉的铝管表面缺陷检测方法因其检测精度高、速度快等优点，已取代人工检测，成为主流检测方法之一。但由于缺陷样本与背景样本之间分布不平衡,导致分类器决策边界偏移、检测精度下降，限制了其应用范围。针对这一问题，提出一种基于集成自适应欠采样的铝管表面缺陷检测方法，首先利用支持向量描述方法对数据分布间的重叠区域进行识别，其次通过构建样本局部密度关系自适应确定欠采样对象及数量，最终利用随机空间生成技术同时对数据样本空间和特征空间进行优化。试验结果表明，所提方法在铝管表面缺陷数据集上识别精确率达到98.52%，优于其他先进检测方法。

关键词: 表面缺陷检测, 支持向量描述, 自适应欠采样, 随机空间生成技术

Abstract: As a standard transmission part, aluminum tube surface defect detection is necessary to ensure product quality and operation safety. The surface defect detection method based on machine vision for aluminum tubes has replaced the manual detection method and become one of the mainstream detection methods due to its metrics such as high detection accuracy and fast detection speed. However, the imbalanced distribution between defective and background samples leads to the shift of the classifier decision boundary, and a decrease in detection accuracy, which limits its application range. To solve this problem, a surface defect detection method of aluminum tube based on ensemble adaptive undersampling is proposed. Firstly, the support vector data description method is used to identify the overlapping areas in the data distribution. Secondly, the local density relationship is utilized to determine which and how many majority samples are to be removed. Finally, the random space generation technology is used to optimize the sample distribution space and feature space simultaneously. The experimental results show that the Precision of the proposed method is 98.52%, which is superior to other advanced detection methods.

Key words: surface defect detection, support vector data description, adaptive undersampling, random space generation

中图分类号:

TP391

郎宁, 王德成, 程鹏. 基于集成自适应欠采样的铝管表面缺陷检测方法研究[J]. 机械工程学报, 2023, 59(6): 18-31.

LANG Ning, WANG Decheng, CHENG Peng. Research on Surface Defect Detection Method of Aluminum Tube Based on Ensemble Adaptive Undersampling[J]. Journal of Mechanical Engineering, 2023, 59(6): 18-31.

参考文献

[1] 中华人民共和国应急管理部. 历史上五月发生的危险化学品事故[EB/OL].[2021-04-29]. https://www.mem.gov.cn/fw/jsxx/202104/t20210429_384168.shtml. Ministry of Emergency Management of the People's Republic of China. History of hazardous chemical accidents in May[EB/OL].[2021-04-29]. https://www.mem.gov.cn/fw/jsxx/202104/t20210429_384168.shtml.
[2] 中华人民共和国应急管理部. 历史上六月发生的危险化学品事故[EB/OL].[2022-06-01]. https://www.mem.gov.cn/fw/jsxx/202206/t20220601_414799.shtml. Ministry of Emergency Management of the People's Republic of China. History of hazardous chemical accidents in June[EB/OL].[2022-06-01]. https://www.mem.gov.cn/fw/jsxx/202206/t20220601_414799.shtml.
[3] 宋乐平,周毅,朱涛. 发动机机油冷却器泄漏分析及改进[J]. 内燃机与动力装置,2017,34(6):44-47,64. SONG Leping,ZHOU Yi,ZHU Tao. Engine oil cooler failure analysis and improvement[J]. Internal Combustion Engine & Powerplant,2017,34(6):44-47,64.
[4] 徐世武,何澍春. 和谐型电力机车复合冷却器泄漏成因分析[J]. 铁道技术监督,2022,50(8):29-34,41. XU Shiwu,HE Shuchun. Analysis on the causes of leakage of compound cooler for harmonious electric locomotive[J]. Railway Quality Control,2022,50(8):29-34,41.
[5] 张仕军,金振林. 基于多模型级联的双目视觉铸件缺陷检测方法[J]. 机械工程学报,2022,58(5):34-43. ZHANG Shijun,JIN Zhenlin. Casting defect detection method based on multi model cascade and binocular vision[J]. Journal of Mechanical Engineering,2022,58(5):34-43.
[6] 胡琨,张树有,赵昕玥,等. 基于Contourlet变换的阴影恢复形貌方法及产品表面缺陷检测应用[J]. 机械工程学报,2018,54(6):102-109. HU Kun,ZHANG Shuyou,ZHAO Xinyue,et al. Defect detection based on contourlet-based shape from shading[J]. Journal of Mechanical Engineering,2018,54(6):102-109.
[7] 周友行,马逐曦,石弦韦,等. HSI颜色空间下的直线导轨表面缺陷检测方法[J]. 中国机械工程,2019,30(18):2179-2184. HU Kun,ZHANG Shuyou,ZHAO Xinyue,et al. Defect detection method of linear guide surfaces based on HSI color space[J]. China Mechanical Engineering, 2019,30(18):2179-2184.
[8] SU B,CHEN H,ZHU Y,et al. Classification of manufacturing defects in multicrystalline solar cells with novel feature descriptor[J]. IEEE Transactions on Instrumentation and Measurement,2019,68(12):4675-4688.
[9] XU C,LI L,LI J,et al. Surface defects detection and identification of lithium battery pole piece based on multi-feature fusion and PSO-SVM[J]. IEEE Access,2021,9:85232-85239.
[10] PARK C,CHOI S,WON S. Vision-based inspection for periodic defects in steel wire rod production[J]. Optical Engineering,2010,49(1):017202-017210.
[11] TAO X,XU D,ZHANG Z,et al. Weak scratch detection and defect classification methods for a large-aperture optical element[J]. Optics Communications,2017,387:390-400.
[12] SAIZ F A,SERRANO I,BARANDIARAN I,et al. A robust and fast deep learning-based method for defect classification in steel surfaces[C]//2018 International Conference on Intelligent Systems (IS),IEEE,Funchal,Portugal,2018:455-460.
[13] YING Z,LIN Z,WU Z,et al. A modified-YOLOv5s model for detection of wire braided hose defects[J]. Measurement,2022,190:110683.
[14] XU H,YAN Z,JI B,et al. Defect detection in welding radiographic images based on semantic segmentation methods[J]. Measurement,2022,188:110569.
[15] LANG N,WANG D,CHENG P,et al. Virtual-sample-based defect detection algorithm for aluminum tube surface[J]. Measurement Science and Technology,2021,32(8):085001.
[16] XIE X,LIU H,ZENG S,et al. A novel progressively undersampling method based on the density peaks sequence for imbalanced data[J]. Knowledge-Based Systems,2021,213:106689.
[17] VUTTIPITTAYAMONGKOL P,ELYAN E. Neighbourhood-based undersampling approach for handling imbalanced and overlapped data[J]. Information Sciences,2020,509:47-70.
[18] ZHENG M,LI T,ZHENG X,et al. UFFDFR:Undersampling framework with denoising,fuzzy c-means clustering,and representative sample selection for imbalanced data classification[J]. Information Sciences,2021,576:658-680.
[19] TSAI C F,LIN W C,HU Y H,et al. Under-sampling class imbalanced datasets by combining clustering analysis and instance selection[J]. Information Sciences,2019,477:47-54.
[20] GALAR M,FERNÁNDEZ A,BARRENECHEA E,et al. EUSBoost:Enhancing ensembles for highly imbalanced data-sets by evolutionary undersampling[J]. Pattern recognition,2013,46(12):3460-3471.
[21] LU W,LI Z,CHU J. Adaptive ensemble undersampling-boost:A novel learning framework for imbalanced data[J]. Journal of systems and software,2017,132:272-282.
[22] TAX D M J,DUIN R P W. Support vector data description[J]. Machine Learning,2004,54(1):45-66.
[23] HO T K. The random subspace method for constructing decision forests[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(8):832-844.
[24] AHMED N,NATARAJAN T,RAO K R. Discrete cosine transform[J]. IEEE transactions on Computers,1974,100(1):90-93.
[25] 杨鑫华,贾昕,朱平,等. 基于信息增益率的点焊接头疲劳性能影响因素分析[J]. 焊接学报,2020,41(10):73-78,101-102. YANG Xinhua,JIA Xin,ZHU Ping,et al. Analysis of factors affecting fatigue performance of welded joints based on information gain rate[J]. Transactions of The China Welding Institution,2020,41(10):73-78,101-102.
[26] VUTTIPITTAYAMONGKOL P,ELYAN E,PETRO-VSKI A. On the class overlap problem in imbalanced data classification[J]. Knowledge-based Systems,2021,212:106631.
[27] DAS S,DATTA S,CHAUDHURI B B. Handling data irregularities in classification:Foundations,trends,and future challenges[J]. Pattern Recognition,2018,81:674-693.
[28] RODRIGUEZ A,LAIO A. Clustering by fast search and find of density peaks[J]. Science,2014,344(6191):1492-1496.
[29] CHAWLA N V,BOWYER K W,HALL L O,et al. SMOTE:Synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research,2002,16:321-357.

基于集成自适应欠采样的铝管表面缺陷检测方法研究

Research on Surface Defect Detection Method of Aluminum Tube Based on Ensemble Adaptive Undersampling

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	宿磊, 王立建, 祁阳, 张思雨, 顾杰斐, 李可. 基于IADSA深度迁移网络的金属表面缺陷检测[J]. 机械工程学报, 2023, 59(24): 46-55.
[2]	吴强威, 文龙. 基于联合对抗训练的深度学习表面缺陷检测方法[J]. 机械工程学报, 2023, 59(12): 173-182.
[3]	李可, 祁阳, 宿磊, 顾杰斐, 苏文胜. 基于改进ACGAN的钢表面缺陷视觉检测方法[J]. 机械工程学报, 2022, 58(24): 32-40.