改进组合分类器的冷轧带钢表面缺陷识别研究

化春键; 周海英

doi:10.13433/j.cnki.1003-8728.2017.1124

改进组合分类器的冷轧带钢表面缺陷识别研究

doi: 10.13433/j.cnki.1003-8728.2017.1124

化春键^1,2,
周海英^1,2

1.
江南大学机械工程学院, 江苏无锡 214122;
2.
江苏省食品先进制造装备技术重点实验室, 江苏无锡 214122

基金项目:

国家自然科学基金项目（61104213）与中央高校基本科研业务费专项资金项目（JUSRP11008）资助

详细信息

作者简介:
化春键(1975-),副教授,博士,cjhua@jiangnan.edu.cn

计量
- 文章访问数: 178
- HTML全文浏览量: 25
- PDF下载量: 5
- 被引次数: 0
出版历程
- 收稿日期: 2016-03-28
- 刊出日期: 2017-11-05

Study on Surface Defect Recognition of Cold Rolled Steel Strip by Improving Combinationclassifier

Hua Chunjian^1,2,
Zhou Haiying^1,2

1.
School of Mechanical Engineering, Jiangnan University, Jiangsu Wuxi 214122, China;
2.
Jiangsu Key Laboratory of Advanced Food Manufacturing Equipment and Technology, Jiangsu Wuxi 214122, China

摘要

摘要: 针对表面缺陷种类多样、形态复杂的冷轧带钢，若采用单一分类器识别分类，会存在对个别缺陷不敏感、识别率低的情况，且会导致分类器处理特征数据规模过大，系统的鲁棒性和稳定性很难得到保证。为此提出基于改进组合分类器的冷轧带钢表面缺陷识别方法，将优化BP神经网络、概率神经网络以及改进的支持向量机进行组合，利用分类信息的互补性进行综合分类，从而构建了较优的分类系统。实验结果表明：改进组合分类器弥补了单个分类器网络训练的不足；针对每一类缺陷识别时准确率都较高，能增加整体分类器的泛化能力，整体识别正确率可达95%以上，且识别高效、稳定，具有实用价值。
- 表面缺陷 /
- 冷轧带钢 /
- 组合分类器 /
- 分类
Abstract: For the cold-rolled strip that has various types of surface defects and complex form, there will be insensitive individual defects and low recognition rate if a single classification is used to identify classification. Therefore it probably results in data classification processing feature size is too large, the robustness and stability of the system is difficult to ensure. The method based on the the improved combination classifier is put forward. It will optimize BP neural network and probabilistic neural network, combining the improved support vector machine. Moreover, it will use complementary classification information to classify. Optimum classification system can be constituted. The experimental results show that the combination of the improved classifier makes up for the lack of a single classifier Network training. For each type of defect recognition, the accuracy is high. It can increase the overall classification generalization ability. The recognition accuracy rate is more than 95%. In short, there is efficient recognition and practical value.
- surface defect /
- cold rolled steel strip /
- combination classifiers /
- classification

HTML全文

参考文献(15)

[1]	张尧,刘伟嵬,邢芝涛,等.采用多分类器集成方法的带钢表面缺陷图像识别[J].东北大学学报(自然科学版),2012,33(2):267-270 Zhang Y, Liu W W, Xing Z T, et al. Surface defect recognition for steel strips by combining multiple classifiers[J]. Journal of Northeastern University(Natural Science), 2012,33(2):267-270(in Chinese)
[2]	韩英莉,洪英.带钢表面缺陷的一种在线检测识别算法研究[J].光电子激光,2015,26(2):320-327 Han Y L, Hong Y. Research on defect surface online detection, classification and recognition algorithm for strip steel[J]. Journal of Optoelectronics·Laser, 2015,26(2):320-327(in Chinese)
[3]	高异,杨延西.基于多支持向量机分类器融合的带钢表面缺陷分类[C]//第三十二届中国控制会议论文集(C卷),西安:西北工业大学自动化学院,2013:3617-3622 Gao Y, Yang Y X. Classification based on multi-classifier of SVM fusion for steel strip surface defects[C]//Proceedings of the 32nd Chinese Control Conference Proceedings(C), Xi'an:Northwestern Polytechnical University Institute of Automation, 2013:3617-3622(in Chinese)
[4]	管声启,王燕妮,师红宇.基于图像预处理的神经网络带钢缺陷检测[J].钢铁研究,2013,41(1):22-26 Guan S Q, Wang Y N, Shi H Y. Neural network inspection of strip defect based on image preprocessing[J]. Research on Iron & Steel, 2013,41(1):22-26(in Chinese)
[5]	Wang X C, Paliwal K K. Feature extraction and dimensionality reduction algorithms and their applications in vowel recognition[J]. Pattern Recognition, 2003,36(10):2429-2439
[6]	Zhang X, Krewet C, Kuhlenkotter B. Automatic classification of defects on the product surface in grinding and polishing[J]. International Journal of Machine Tools and Manufacture, 2006,46(1):59-69
[7]	王小川,史峰,郁磊,等.MATLAB神经网络43个案例分析[M].北京:北京航空航天大学出版社,2013 Wang X C, Shi F, Yu L, et al. MATLAB neural network 43 case analysis[M]. Beijing:Beihang University Press, 2013(in Chinese)
[8]	翟越,刘浪,于澍.堤防管涌发生可能性识别的网格搜索-支持向量机方法[J].中南大学学报(自然科学版),2015,46(4):1497-1503 Zhai Y, Liu L, Yu S. Utilization of nonlinear SVM with grid-search method for identification of piping occurring probability in embankment engineering[J]. Journal of Central South University(Science and Technology), 2015,46(4):1497-1503(in Chinese)
[9]	Suzuki R, Kawai F, Nakazawa C, et al. Parameter optimization of model predictive control by PSO[J]. Electrical Engineering in Japan, 2012,178(1):40-49
[10]	Guo H S, Wang W J. An active learning-based SVM multi-class classification model[J]. Pattern Recognition, 2015,48(5):1577-1597
[11]	Li S F. BP alogrithm in pattern recognition of glass defects[C]//Proceedings of 2012 IEEE 5th International Conference on Advanced Computational Intelligence, October 18-20, 2012, Nanjing, China. Nanjing:IEEE, 2012:183-187
[12]	蔡英凤,王海.视觉车辆识别迁移学习算法[J].东南大学学报(自然科学版),2015,45(2):275-280 Cai Y F, Wang H. Vision based vehicle detection transfer learning algorithm[J]. Journal of Southeast University(Natural Science Edition), 2015,45(2):275-280(in Chinese)
[13]	Hammer B, Villmann T. Generalized relevance learning vector quantization[J]. Neural Networks, 2002,15(8-9):1059-1068
[14]	王小虎,张石清,曹恒瑞.基于多分类器集成的语音情感识别[J].微电子学与计算机,2015,32(7):38-41,45 Wang X H, Zhang S Q, Cao H R. Speech emotion recognition based on integration of multiple classifiers[J]. Microelectronics & Computer, 2015,32(7):38-41,45(in Chinese)
[15]	周新星,王典洪,孙林.基于独立成分分析的表面缺陷特征提取与识别方法[J].计算机辅助设计与图像学学报,2012,24(4):506-513 Zhou X X, Wang D H, Sun L. Feature extraction and recognition method of surface defects based on independent component analysis[J]. Journal of Computer-Aided Design & Computer Graphics, 2012,24(4):506-513(in Chinese)