Please wait a minute...
浙江大学学报(工学版)  2020, Vol. 54 Issue (10): 1929-1935    DOI: 10.3785/j.issn.1008-973X.2020.10.009
计算机技术     
变照度下的视觉测量系统误差建模
粟序明1(),方成刚1,2,*(),潘裕斌1,吴伟伟3,李亚萍4,朱浪1
1. 南京工业大学 机械与动力工程学院,江苏 南京 210000
2. 南京工业大学 江苏省工业装备数字制造及控制技术重点实验室,江苏 南京 210009
3. 扬州大学 机械工程学院,江苏 扬州 225000
4. 华东理工大学 化学与分子工程学院,上海 200237
Modeling error of visual measurement system under changing illuminance
Xu-ming SU1(),Cheng-gang FANG1,2,*(),Yu-bin PAN1,Wei-wei WU3,Ya-ping LI4,Lang ZHU1
1. School of Mechanical and Power Engineering, Nanjing University of Technology, Nanjing 210000, China
2. Jiangsu Key Laboratory of Industrial Equipment Digital Manufacturing and Control Technology, Nanjing University of Technology, Nanjing 210009, China
3. College of Mechanical Engineering, Yangzhou University, Yangzhou 225000, China
4. School of Chemistry and Molecular Engineering, East China University of Science and Technology, Shanghai 200237, China
 全文: PDF(1169 KB)   HTML
摘要:

针对二维视觉在线测量工件时,照度变化因素导致测量误差的问题,提出基于遗传算法优化的最小二乘支持向量机(GA-LSSVM),建立照度误差模型的方法. 分析视觉测量系统的误差来源,通过最小二乘法分析照度影响下的误差规律. 利用照度变化误差实验,获得照度和测量系统的误差数据,分别训练GA-LSSVM、支持向量机(SVM)以及BP神经网络,建立照度和测量系统误差模型,对系统测量误差进行预测. 结果表明:在变照度测量误差预测模型中,GA-LSSVM模型、SVM模型及BP神经网络模型的预测精度分别为94.90%、90.23%及80.60%. 这表明遗传算法优化的最小二乘支持向量机建立的变照度误差模型,在拟合和预测精度上优于传统的BP神经网络.

关键词: 视觉测量误差建模遗传算法优化的最小二乘支持向量机照度BP神经网络    
Abstract:

A modeling method of the illuminance error was proposed based on genetic algorithm (GA) to optimize the least squares support vector machine (LSSVM) aiming at the problem that illuminance changes led to the error of measurement for online measuring two-dimensional visual artifacts. Vision measurement system error sources were analyzed, and illuminance under the influence was analyzed by the least squares method of error patterns. The illuminance and the error of the measurement system data were obtained by using the illumination change error of experiment. The illuminance and measurement system error model was established by training the GA - LSSVM, support vector machine (SVM) and BP neural network respectively to forecast the system measurement error. Results show that the prediction accuracy of GA-LSSVM model, SVM model and BP neural network model is 94.90%, 90.23% and 80.60% respectively in the prediction model of variable illuminance measurement error. The error model of variable illuminance that is established by the least square support vector machine and optimized by genetic algorithm is superior to the traditional BP neural network in terms of fitting and prediction accuracy.

Key words: visual measurement    error modeling    genetic algorithm optimization least squares support vector machine    illuminance    BP neural network
收稿日期: 2019-08-28 出版日期: 2020-10-28
CLC:  TP 391  
基金资助: 国家自然科学基金资助项目(51635003);江苏省科技成果转化专项资金资助项目(BA2017099)
通讯作者: 方成刚     E-mail: 983438713@qq.com;279119134@qq.com
作者简介: 粟序明(1994—),男,硕士生,从事机器视觉高精密测量及智能装备研究. orcid.org/0000-0003-3717-9940. E-mail: 983438713@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  
粟序明
方成刚
潘裕斌
吴伟伟
李亚萍
朱浪

引用本文:

粟序明,方成刚,潘裕斌,吴伟伟,李亚萍,朱浪. 变照度下的视觉测量系统误差建模[J]. 浙江大学学报(工学版), 2020, 54(10): 1929-1935.

Xu-ming SU,Cheng-gang FANG,Yu-bin PAN,Wei-wei WU,Ya-ping LI,Lang ZHU. Modeling error of visual measurement system under changing illuminance. Journal of ZheJiang University (Engineering Science), 2020, 54(10): 1929-1935.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2020.10.009        http://www.zjujournals.com/eng/CN/Y2020/V54/I10/1929

图 1  视觉测量系统误差影响因素来源
图 2  测量误差与3个位置照度总和的关系
图 3  遗传算法变照度误差补偿模型方法
图 4  遗传算法优化的最小二乘支持向量机变照度误差补偿模型方法
图 5  变照度误差检测实验平台
图 6  实验轴提取的边缘
图 7  原图上拟合的边缘
图 8  3个不同位置照度计照度数据
图 9  不同照度下的原始误差数据
图 10  遗传算法优化的最小二乘支持向量机变照度误差模型预测结果
图 11  支持向量机变照度误差模型预测结果
图 12  BP神经网络变照度误差模型预测结果
模型 $\varDelta _{\rm{max} }/{ { {\text{μm} }} }$ ${\gamma }_{\rm{max} }/{\text{μm} }$ $\bar\varDelta /{\text{μm} }$ $ \delta / \text{%} $
遗传算法优化的最小二乘支持向量机变照度误差预测模型 114.3 5.83 2.19 94.90
支持向量变照度误差预测模型 114.3 10.92 4.11 90.50
BP神经网络变照度误差
预测模型
114.3 22.14 5.67 80.60
表 1  变照度误差预测模型精度对比
1 AHN S J, WARNECKE H J, KOTOWSKI R Systematic geometric image measurement errors of circular object targets: mathematical formulation and correction[J]. The Photogrammetric Record, 1999, 16 (93): 485- 502
doi: 10.1111/0031-868X.00138
2 CUI J W, TAN J B, ZHOU Y, et al Improvement of vision measurement accuracy using Zernike moment-based edge location error compensation model[J]. Journal of Physics: Conference Series, 2007, 48 (1): 1353
3 杨君, 张涛, 宋靖雁, 等 星点质心亚像元定位的高精度误差补偿法[J]. 光学精密工程, 2010, 18 (4): 1002- 1010
YANG Jun, ZHANG Tao, SONG Jing-yan, et al High precision error compensation method for sub-pixel positioning of star point centroid[J]. Optics and Precision Engineering, 2010, 18 (4): 1002- 1010
4 YANG S H, NATARAJAN U, SEKAR M, et al Prediction of surface roughness in turning operations by computer vision using neural network trained by differential evolution algorithm[J]. International Journal of Advanced Manufacturing Technology, 2010, 51 (9-12): 965- 971
doi: 10.1007/s00170-010-2668-5
5 邾继贵, 邹剑, 林嘉睿, 等 摄影测量图像处理的高精度误差补偿法[J]. 光学学报, 2012, 32 (9): 129- 136
ZHU Ji-gui, ZOU Jian, LIN Jia-rui, et al High-precision error compensation method for photogrammetric image processing[J]. Acta Optica Sinica, 2012, 32 (9): 129- 136
6 ZHU W D, MEI B, YAN G R, et al Measurement error analysis and accuracy enhancement of 2D vision system for robotic drilling[J]. Robotics and Computer-Integrated Manufacturing, 2014, 30 (2): 160- 171
doi: 10.1016/j.rcim.2013.09.014
7 WANG J M, GAO B, ZHANG X, et al. Error correction for high-precision measurement of cylindrical objects diameter based on machine vision [C]// Proceedings of 2015 IEEE 12th International Conference on Electronic Measurement and Instruments. Qingdao: IEEE, 2015: 1113-1117.
8 DEND H X, WANG F, ZHANG J, et al Vision measurement error analysis for nonlinear light refraction at high temperature[J]. Applied Optics, 2018, 57 (20): 5556- 5565
doi: 10.1364/AO.57.005556
9 LI K, YUAN F, DING Z L, et al. Vision measurement error compensation research of double-theodolite based on neural network approaching [C]// Control Conference on IEEE Control Systems. [S. l.]: IEEE, 2011: 2815-2820.
10 SHEN Y J, ZHANG X, CHENG W, et al Quasi-eccentricity error modeling and compensation in vision metrology[J]. Measurement Science and Technology, 2018, 29 (4): 045006
11 LIU W, MA X, LI X, et al A novel vision-based pose measurement method considering the refraction of light[J]. Sensors, 2018, 18 (12): 4348- 4363
doi: 10.3390/s18124348
12 KLOSOWSKI M, JAKUSZ J A high-efficient measurement system with optimization feature for prototype CMOS image sensors[J]. IEEE Transactions on Instrumentation and Measurement, 2018, 67 (10): 2363- 2372
doi: 10.1109/TIM.2018.2814118
13 杜文华, 郭小刚, 曾志强, 等 自适应光强变化的动态边缘位置提取方法[J]. 包装工程, 2018, 39 (3): 162- 166
DU Wen-hua, GUO Xiao-gang, ZENG Zhi-qiang, et al Dynamic edge position extraction method for adaptive light intensity variation[J]. Packaging Engineering, 2018, 39 (3): 162- 166
[1] 黄智,贾臻杰,邓涛,刘永超,杜丽. 基于支持向量机的静压转台热误差补偿[J]. 浙江大学学报(工学版), 2019, 53(8): 1594-1601.
[2] 丰明坤,施祥. 视觉特征深度融合的图像质量评价[J]. 浙江大学学报(工学版), 2019, 53(3): 512-521.
[3] 谢杰, 黄筱调, 方成刚, 周宝仓, 陆宁. 磨齿机电主轴热特性及热误差建模[J]. 浙江大学学报(工学版), 2018, 52(2): 247-254.
[4] 孟濬, 邓晓雨, 虞捷舟. 基于变量聚类的BP神经网络术后生存期预测模型[J]. 浙江大学学报(工学版), 2018, 52(12): 2365-2371.
[5] 尤海辉, 马增益, 唐义军, 王月兰, 郑林, 俞钟, 吉澄军. 循环流化床入炉垃圾热值软测量[J]. 浙江大学学报(工学版), 2017, 51(6): 1163-1172.
[6] 毕运波, 徐超, 樊新田, 严伟苗. 基于视觉测量的沉头孔垂直度检测方法[J]. 浙江大学学报(工学版), 2017, 51(2): 312-318.
[7] 巫江虹, 薛志强, 金鹏, 李会喜. 电动汽车热泵空调微通道换热器温度分布特性[J]. 浙江大学学报(工学版), 2016, 50(8): 1537-1544.
[8] 张阿龙, 章明, 乔明杰, 朱伟东, 梅标. 基于视觉测量的环形轨底座位姿标定方法[J]. 浙江大学学报(工学版), 2016, 50(6): 1080-1087.
[9] 刘磊,杨鹏,刘作军. 基于多源信息和粒子群优化算法的下肢运动模式识别[J]. 浙江大学学报(工学版), 2015, 49(3): 439-447.
[10] 黄晓烁,何衍,蒋静坪. 基于互联网无刷直流电机传动系统的控制策略[J]. J4, 2013, 47(5): 831-836.
[11] 童水光, 王相兵, 钟崴, 张健. 基于BP-HGA的起重机刚性支腿动态优化设计[J]. J4, 2013, 47(1): 122-130.
[12] 王德明, 王莉, 张广明. 基于遗传BP神经网络的短期风速预测模型[J]. J4, 2012, 46(5): 837-841.
[13] 赵振, 张树有. 全分断周期混合电流模型及其应用[J]. J4, 2012, 46(2): 301-308.
[14] 徐敬华, 张树有. 基于形态分布图与BP神经网络的三维模型检索方法[J]. J4, 2009, 43(5): 877-883.
[15] 宓云軿 王晓萍 金鑫. 基于机器学习的水质COD预测方法[J]. J4, 2008, 42(5): 790-793.