[1]陈 跃.改进可拓理论的带钢表面缺陷图像分类方法[J].南京师范大学学报(工程技术版),2016,16(03):054.[doi:10.3969/j.issn.1672-1292.2016.03.009]
 Chen Yue.Classification of Steel Strip Surface Defect ImagesBased on Improved Extenics Theory[J].Journal of Nanjing Normal University(Engineering and Technology),2016,16(03):054.[doi:10.3969/j.issn.1672-1292.2016.03.009]
点击复制

改进可拓理论的带钢表面缺陷图像分类方法
分享到:

南京师范大学学报(工程技术版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
16卷
期数:
2016年03期
页码:
054
栏目:
计算机工程
出版日期:
2016-09-30

文章信息/Info

Title:
Classification of Steel Strip Surface Defect ImagesBased on Improved Extenics Theory
文章编号:
1672-1292(2016)03-0054-09
作者:
陈 跃
徐州工程学院机电工程学院,江苏 徐州 221111
Author(s):
Chen Yue
School of Mechanical & Electrical Engineering,Xuzhou Institute of Technology,Xuzhou 221111,China
关键词:
带钢缺陷图像可拓理论关联度缺陷分类
Keywords:
steel strip defect imageextenics theoryrelevancy degreedefects classification
分类号:
TP391
DOI:
10.3969/j.issn.1672-1292.2016.03.009
文献标志码:
A
摘要:
将可拓理论引入到带钢表面缺陷图像的分类中,提取分割前后缺陷图像的12个特征值,对可拓理论用于分类的关键步骤-关联度的计算方法进行改进,首先确定某缺陷特征值与某类缺陷对应特征值经典域的距绝对值,再计算出该特征值与各类缺陷对应特征值经典域的距绝对值之和,以二者的商作为关联度计算的权值. 对该方法的可行性进行了论证,并对带钢表面缺陷图像分类进行了仿真,仿真结果显示改进后的方法分类准确性有较大的提高.
Abstract:
Extenics theory is introduced into steel strip defects’ images classification. Twelve features are extracted from segmented and unsegmented images. The key step-relevancy values computering method is improved. The quotients of distance between feature values and classical domain and sum of these distances are used as weight coffecients in computering comprehensive relevancy values. This computering method enhances the influences of defects’ self feature values on comprehensive relevancy. The effectiveness of the improved computering method is also demonstrated. Steel strip defects are selected to simulate the method,maximum relevancy value is used to group the unspecified defect image in one of preselected defect types,comparing to primary weight coefficient computering method,the improved theory is more effective in defects images classification.

参考文献/References:

[1] 丁世飞,齐丙娟,谭红艳. 支持向量机理理论与算法研究综述[J]. 电子科技大学学报,2011,40(1):2-9.
DING S F,QI B J,TAN H Y. An overview on theory and algorithm of support vector machines[J]. Journal of university of electronic science and technology of China,2011,40(1):2-9. (in Chinese)
[2] SAURABH P,MALIK M I,PETROS D. Feature selection for linear SVM with provable guarantees[J]. Pattern?recognition,2016,60:205-214.
[3] 储茂祥,王安娜,巩荣芬. 一种改进的最小二乘孪生支持向量机分类算法[J]. 电子学报,2014,42(5):998-1 003.
CHU M X,WANG A,GONG R F. Improvement on least squares twin support vector machine for pattern classification[J]. Chinese journal of electronics,2014,42(5):998-1 003. (in Chinese)
[4] 刘绍毓,周杰,李弼程,等. 基于多分类SVM-KNN的实体关系抽取方法[J]. 数据采集与处理,2015,1(30):202-210.
LIU S Y,ZHOU J,LI B C,et al. Entity relation extraction method based on multi-SVM-KNN classifier[J]. Journal of data acquisition and processing,2015,1(30):202-210. (in Chinese)
[5] 胡学坤,李金霞. 基于粗糙集与模糊支持向量机的模式分类方法研究[J]. 科技通报,2010,26(2):250-252.
HU X K,LI J X. Method of pattern classification based on FSVM and RS theory[J]. Bulletin of science and technology,2010,26(2):250-252. (in Chinese)
[6] 陈方林,刘彦. 基于支持向量机的X射线焊缝缺陷检测[J]. 机械工程与自动化,2010(2):122-126.
CHEN F L,LIU Y. Defect detection of X-ray image of weld using support vector machine[J]. Mechanical engineering & automation,2010(2):122-126. (in Chinese)
[7] 王再超,李光辉,冯海林,等. 基于应力波和支持向量机的木材缺陷识别分类方法[J]. 南京林业大学学报(自然科学版),2015,39(3):130-136.
WANG Z C,LI G H,FENG H L,et al. A method of wood defect identification and classification based on stress wave and SVM[J]. Journal of Nanjing forestry university(natural sciences edition),2015,39(3):130-136. (in Chinese)
[8] 黎维娟,卢振泰. 基于支持向量机的脑部MR图像细分类[J]. 电路与系统学报,2010,15(1):5-9.
LI W J,LU Z T,FENG Q J,et al. MR-brain image meticulous classification based on support vector machine[J]. Journal of circuits and systems,2010,15(1):5-9. (in Chinese)
[9] JORDI I. Automatic recognition of man-made objects in high resolution optical remote sensing images by SVM classification of geometric image features[J]. ISPRS journal of photogrammetry and remote sensing,2007,62(3):236-248.
[10] Anagnostopoulos G C. SVM-based target recognition from synthetic aperture radar images using target region outline descriptors[J]. Nonlinear analysis theory methods & applications,2009,71(12):e2934-e2939.
[11] 蔡文. 可拓集合和不相容问题[J]. 科学探索学报,1983(1):83-97.
CAI W. Extension set and non-compatible problems[J]. Journal of scientific exploration,1983(1):83-97. (in Chinese)
[12] 杨春燕,蔡文. 可拓工程[M]. 北京:科学出版社,2007.
YANG C Y,CAI W. Extension engineering[M]. Beijing:Science Press,2007. (in Chinese)
[13] WANG M H,TSENG Y F. A novel clustering algorithm based on the extension theory and genetic algorithm[J]. Expert systems with applications,2009,36(4):8 269-8 276.
[14] 刘海,朱小平. 一种基于可拓理论的图像检索方法[J]. 计算机系统应用,2009(3):54-56.
LIU H,ZHU X P,XIA M B. An image retrieval method based on cotorgy[J]. Application of computer system,2009(3):54-56. (in Chinese)
[15] AILING C,LIPING L,XINGSEN L,et al. Study on innovation capability of college students based on extenics and theory of creativity[J]. Procedia computer science,2013,17:1 194-1 201.
[16] JIAMIN W,YE T,Mindan L,et al. Analysis on test cheating and its solutions based on extenics and information technology[J]. Procedia computer science,2015,55:1 009-1 014.
[17] 张家宾,张金春,李日华,等. 基于可脱学的故障诊断及预防方法研究[J]. 广东工业大学学报,2015,1(32):11-15.
ZHANG J B,ZHANG J C,LI R H,et al. Research on fault diagnosis and prevention based on extension[J]. Journal of Guangdong university of technology,2015,1(32):11-15. (in Chinese)
[18] HU M. Visual pattern recognition by moment invariants[J]. Information theory ire transactions on,1962,8(2):179-187.

备注/Memo

备注/Memo:
收稿日期:2016-04-11. 
基金项目:江苏省高校自然科学基金(10KJD510010). 
通讯联系人:陈跃,博士,副教授,研究方向:图像检测,自动控制. E-mail:snake9521@163.com
更新日期/Last Update: 2016-09-30