Please wait a minute...
工程设计学报  2022, Vol. 29 Issue (4): 410-418    DOI: 10.3785/j.issn.1006-754X.2022.00.043
设计理论与方法     
一种零件综合质量评定方法研究
李梦(),尹宗军
安徽信息工程学院 机械工程学院,安徽 芜湖 241000
Research on comprehensive quality assessment method for parts
Meng LI(),Zong-jun YIN
Department of Mechatronics Engineering, Anhui Institute of Information Technology, Wuhu 241000, China
 全文: PDF(4670 KB)   HTML
摘要:

零件的质量评定是柔性智能制造中十分重要的环节。现有的自动化识别装置一般采用非人工接触的光学检测系统,但由于工况环境复杂,诸多干扰因素均会影响零件质量检测与评定的准确性。另外,工业现场的连续作业对工控机硬件的运行速度、光学检测系统的环境适应性以及质量评定算法的预测准确性都提出了更高的要求。基于此,提出一种基于机器视觉与机器学习的零件综合质量评定方法。首先,借助机器视觉技术完成被测零件图像的实时采集与处理,并利用灰度匹配算法与几何匹配算法对零件的图像与CAD(computer aided design,计算机辅助设计)机械加工图进行比较,求解得到灰度匹配分数与几何匹配分数这2个几何特征参数。然后,针对零件表面的缺陷(如划伤、磨损、边缘缺料及锈蚀等),在图像预处理(灰度化、图像增强、高斯降噪和二值化)的基础上,求解得到图像灰度的均值和标准差这2个表面缺陷特征参数。最后,借助主成分分析(principal component analysis, PCA)对零件的四维特征数据集进行降维处理,并利用K最近邻(K-nearest neighbor, KNN)算法对降维后的数据集进行训练和预测,完成零件综合质量评定;在此基础上,比较KNN算法与其他机器学习算法的准确率、召回率和特异度等指标,以验证其可行性。实验结果表明,所搭建的光学检测与处理系统在不同光源条件下的识别准确率达到96.15%以上;当相机的快门时间设定为100 μs时,该系统的图像处理速度达到45.2 帧/s。所提出的零件综合质量评定方法具有较高的准确率与处理速度,适用于复杂工况下零件的综合质量评定。

关键词: 质量评定几何匹配灰度匹配表面缺陷特征提取机器学习    
Abstract:

The quality assessment of parts is a very important link in flexible intelligent manufacturing. The existing automatic identification device generally adopts the non-artificial contact optical detection system, but due to the complex working environment, many interference factors affect the accuracy of quality detection and assessment for parts. In addition, the continuous operation of industrial site puts forward higher requirements on the running speed of industrial control machine hardware, the environmental adaptability of optical detection system and the prediction accuracy of part quality assessment algorithm. Based on this, a comprehensive quality assessment method for parts based on the machine vision and machine learning was proposed. Firstly, the real-time acquisition and processing of the measured part image was completed with the help of machine vision technology, and then the gray matching algorithm and geometric matching algorithm were used to compare the images and the CAD (computer aided design) machining drawings of parts, so as to solve the geometric feature parameters of gray matching score and the geometric matching score. Then, according to the surface defects of parts (such as scratch, wear, edge material shortage and rust), the surface defect feature parameters of mean and standard deviation of image gray were solved on the basis of image pretreatment (gray, image enhancement, Gaussian noise reduction and binary). Finally, the four-dimensional feature data set of parts was dimensionally reduced by the principal component analysis (PCA), and the K-nearest neighbor (KNN) algorithm was used to train and predict the data set after dimension reduction to complete the comprehensive quality assessment for parts. On this basis, the accuracy, recall rate, specificity and other indicators of the KNN algorithm and other machine learning algorithms were compared to verify its feasibility. The experimental results showed that, the recognition accuracy of optical detection and processing system was more than 96.15% under different lighting conditions; when the camera shutter time was set to 100 μs, the image processing speed of this system reached 45.2 frames/s. The proposed comprehensive quality assessment method for parts has high accuracy and processing speed, which is suitable for comprehensive quality assessment of parts under complex working conditions.

Key words: quality assessment    geometric matching    gray matching    surface defect feature extraction    machine learning
收稿日期: 2021-08-30 出版日期: 2022-09-05
CLC:  TH-39  
基金资助: 安徽省教育厅自然科学研究项目(KJ2019A1285)
作者简介: 李 梦(1986—),女,安徽淮南人,副教授,硕士,从事机器视觉与智能控制研究,E-mail:menlikelye@163.comhttps://orcid.org/0000-0003-4756-6653
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
李梦
尹宗军

引用本文:

李梦,尹宗军. 一种零件综合质量评定方法研究[J]. 工程设计学报, 2022, 29(4): 410-418.

Meng LI,Zong-jun YIN. Research on comprehensive quality assessment method for parts[J]. Chinese Journal of Engineering Design, 2022, 29(4): 410-418.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn.1006-754X.2022.00.043        https://www.zjujournals.com/gcsjxb/CN/Y2022/V29/I4/410

图1  零件综合质量评定方法总体框架
图2  边缘点灰度梯度方向角示意
图3  光学检测与处理系统
图4  零件的CAD机械加工图
图5  零件图像的模板匹配结果
图6  存在不同表面缺陷的零件图像预处理结果
图7  零件的四维特征数据集及其综合质量评定结果(部分)
图8  基于PCA+KNN算法的零件综合质量分类结果
图9  基于不同算法的零件综合质量分类结果对比(几何匹配分数—灰度匹配分数)
图10  基于不同算法的零件综合质量分类结果对比(灰度匹配分数—灰度标准差)
图11  基于决策树的零件综合质量分类结果
最小样本叶取值准确率/%
396.37
893.76
2087.43
表1  基于决策树的零件综合质量分类准确率对比
指标PCA+KNN算法K均值聚类算法异常点检测算法决策树
准确率94.29~97.4193.3192.7787.43~96.37
召回率92.53~95.2393.7893.6291.98~95.28
特异度92.12~95.9292.7594.3491.43~94.87
精准率93.45~95.7694.1793.4292.33~94.61
F1分数92.34~94.2292.8393.7393.37~95.05
表2  基于不同算法的零件综合质量分类结果比较 (%)
光源照度/lx快门时间/μs处理速度/(帧/s)准确率/%
50010045.297.41
30015041.696.83
20020038.496.15
表3  不同光源条件下基于PCA+KNN算法的零件综合质量分类准确率
1 LIU Q Z, SUNG A H, CHEN Z X, et al. Feature mining and pattern classification for steganalysis of LSB matching steganography in grayscale images[J]. Pattern Recognition, 2008, 41(1): 56-66. doi:10.1016/j.patcog.2007.06.005
doi: 10.1016/j.patcog.2007.06.005
2 JIA Xi-yue, CAO Yi-ning, David O’CONNOR, et al. Mapping soil pollution by using drone image recognition and machine learning at an arsenic-contaminated agricultural field[J]. Environmental Pollution, 2021, 270: 116281. doi:10.1016/j.envpol.2020.116281
doi: 10.1016/j.envpol.2020.116281
3 WEI Wei, DING Lie-yun, LUO Han-bin, et al. Automated bughole detection and quality performance assessment of concrete using image processing and deep convolutional neural networks[J]. Construction and Building Materials, 2021, 281: 122576. doi:10.1016/j.conbuildmat.2021.122576
doi: 10.1016/j.conbuildmat.2021.122576
4 蓝金辉,王迪,申小盼.卷积神经网络在视觉图像检测的研究进展[J].仪器仪表学报,2020,41(4):167-182. doi:10.19650/j.cnki.cjsi.J2006003
LAN Jin-hui, WANG Di, SHEN Xiao-pan. Research progress on visual image detection based on convolutional neural network[J]. Chinese Journal of Scientific Instrument, 2020, 41(4): 167-182.
doi: 10.19650/j.cnki.cjsi.J2006003
5 SAHA Dhritiman, MANICKAVASAGAN Annamalai. Machine learning techniques for analysis of hyperspectral images to determine quality of food products: a review[J]. Current Research in Food Science, 2021, 4: 28-44. doi:10.1016/j.crfs.2021.01.002
doi: 10.1016/j.crfs.2021.01.002
6 AYESHA, QURESHI M B, AFZAAL M, et al. Machine learning-based EEG signals classification model for epileptic seizure detection[J]. Multimedia Tools and Applications, 2021, 80(12): 17849-17877. doi:10.1007/s11042-021-10597-6
doi: 10.1007/s11042-021-10597-6
7 赵勇,李怀宇.基于通用距离测量的机器学习方法用于图像分类和聚类[J].电子测量技术,2017,40(9):136-140. doi:10.3969/j.issn.1002-7300.2017.09.026
ZHAO Yong, LI Hai-yu. Machine learning methods based on universal distance measurement for image classification and clustering[J]. Electronic Measurement Technology, 2017, 40(9): 136-140.
doi: 10.3969/j.issn.1002-7300.2017.09.026
8 吴春晓,行鸿彦,张漪俊.基于BP神经网络的地温推演模型[J].电子测量与仪器学报,2017,31(10):1561-1567.
WU Chun-xiao, HANG Hong-yan, ZHANG Yi-jun. Ground temperature deduction model based on BP neural network[J]. Journal of Electronic Measurement and Instrumentation, 2017, 31(10): 1561-1567.
9 陶显,侯伟,徐德.基于深度学习的表面缺陷检测方法综述[J].自动化学报,2021,47(5):1017-1034. doi:10.16383/j.aas.c190811
TAO Xian, HOU Wei, XU De. A survey of surface defect detection methods based on deep learning[J]. Acta Automatica Sinica, 2021, 47(5): 1017-1034.
doi: 10.16383/j.aas.c190811
10 杨明莉,范玉刚,李宝芸.基于LDA和ELM的高光谱图像降维与分类方法研究[J].电子测量与仪器学报,2020,34(5):190-196. doi:10.13382/j.jemi.B1902756
YANG Ming-li, FAN Yu-gang, LI Bao-yun. Research on dimensionality reduction and classification of hyperspectral images based on LDA and ELM[J]. Journal of Electronic Measurement and Instrument, 2020, 34(5): 190-196.
doi: 10.13382/j.jemi.B1902756
11 SUN Wei-chen, ZHANG Zhi-jing, SHI Ling-ling, et al. Small sample parts recognition and localization from unfocused images in precision assembly systems using relative entropy[J]. Precision Engineering, 2021, 68: 206-217. doi:10.1016/j.precisioneng.2020.12.015
doi: 10.1016/j.precisioneng.2020.12.015
12 UCEV A, MINALINEE T T. Machine parts recognitionand defect detection in automated assembly systems using computer vision techniques[J]. Revista Técnica De La Facultad De Ingeniería Universidad Del Zulia, 2016, 39(1): 71-80.
13 BOHLOOL M, TAGHANAKI S R. Cost-efficient automated visual inspection system for small manufacturing industries based on SIFT[C]//23rd International Conference Image and Vision Computing, Christchurch, New Zealand, Nov. 26-28, 2008.
14 PICCININI P, PRATI A, CUCCHIARA R. Real-time object detection and localization with SIFT-based clustering[J]. Image and Vision Computing, 2012, 30(8): 573-587. doi:10.1016/j.imavis.2012.06.004
doi: 10.1016/j.imavis.2012.06.004
15 RASHAD M Z, ELDESOUKY B S, KHAWASIK M S. Plants images classification based on textural features using combined classifier[J]. International Journal of Computer Science and Information Technology, 2011, 3(4): 93-100. doi:10.5121/ijcsit.2011.3407
doi: 10.5121/ijcsit.2011.3407
16 ALAMOUDI S, HONG X, WEI H. Plant leaf recognitionusing texture features and semi-supervised spherical k-means clustering[C]//International Joint Conference on Neural Networks (IJCNN), Glasgow, Jul. 19-24, 2020. doi:10.1109/ijcnn48605.2020.9207386
doi: 10.1109/ijcnn48605.2020.9207386
17 刘丽,赵凌君,郭承玉,等.图像纹理分类方法研究进展和展望[J].自动化学报,2018,44(4):584-607. doi:10.16383/j.aas.2018.c160452
LIU Li, ZHAO Ling-jun, GUO Cheng-yu, et al. Texture classification: state-of-the-art methods and prospects[J]. Acta Automatica Sinica, 2018, 44(4): 584-607.
doi: 10.16383/j.aas.2018.c160452
18 李晖,吴佳宁,苑玮琦,等.基于视觉显著性的木板实时分类方法研究[J].仪器仪表学报,2018,39(12):237-244. doi:10.19650/j.cnki.cjsi.J1803700
LI Hui, WU Jia-ning, YUAN Wei-qi, et al. Real-time classification of wood based on visual significance[J]. Chinese Journal of Scientific Instrument, 2018, 39(12): 237-244.
doi: 10.19650/j.cnki.cjsi.J1803700
19 YANG Cheng-zhuan, YU Qian. Multiscale fourier descriptor based on triangular features for shape retrieval[J]. Signal Processing Image Communication, 2019, 71: 110-119. doi:10.1016/j.image.2018.11.004
doi: 10.1016/j.image.2018.11.004
20 LEE S H, CHAN C S, REMAGNINO P. Multi-organ plant classification based on convolutional and recurrent neural networks[J]. IEEE Transactions on Image Processing, 2018, 27(9): 4287-4301. doi:10.1109/tip.2018.2836321
doi: 10.1109/tip.2018.2836321
21 杨高科.图像处理、分析与机器视觉[M].北京:清华大学出版社,2018:472-485.
YANG Gao-ke. Image processing, analysis, and machine vision[M]. Beijing: Tsinghua University Press, 2018: 472-485.
22 周志华.机器学习[M].北京:清华大学出版社,2018:229-232.
ZHOU Zhi-hua. Machine learning[M]. Beijing: Tsinghua University Press, 2018: 229-232.
[1] 谢红太,王红,柴伟. 新型高速列车风阻制动装置设计与仿真分析[J]. 工程设计学报, 2023, 30(2): 244-253.
[2] 严颖, 张小平, 姜海鹏, 张铸, 赵延明, 黄良沛. 基于GSSEC的开关磁阻电机驱动的海洋绞车主动升沉补偿控制方法[J]. 工程设计学报, 2021, 28(2): 132-140.
[3] 王奔, 朱龙彪, 沈祖军, 陈小林. LYP1050轮转胶印机控制系统设计[J]. 工程设计学报, 2021, 28(1): 112-120.
[4] 曹鹏勇, 王建文. 基于STM8S105的智能车结构及控制系统的研究[J]. 工程设计学报, 2020, 27(4): 516-523.
[5] 穆载乐, 方娟, 陈隆飞, 张秋菊. 上下肢康复机器人的结构与控制系统设计[J]. 工程设计学报, 2019, 26(6): 736-742.
[6] 杨蕾, 任成祖. 圆柱滚子电磁推进装置仿真与推进实验分析[J]. 工程设计学报, 2019, 26(5): 611-618.