基于超熵和模糊集理论的带钢表面缺陷分割

杨永敏; 樊继壮; 赵杰

doi:10.3788/OPE.20111907.1651

您当前的位置：

首页 >

文章列表页 >

基于超熵和模糊集理论的带钢表面缺陷分割

信息科学 | 更新时间：2020-08-12

- 基于超熵和模糊集理论的带钢表面缺陷分割
- Steel strip surface defect segmentation based on excess entropy and fuzzy set theory
- 光学精密工程 2011年19卷第7期页码：1651-1658
- 作者机构：
  
  哈尔滨工业大学机器人技术与系统国家重点实验室,黑龙江哈尔滨,150080
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2007AA041501)();国家重大科技专项资助项目(No.2009ZX04004-062)();哈尔滨工业大学科研创新
- DOI：10.3788/OPE.20111907.1651
  中图分类号： TP391.4
- 收稿日期：2010-09-25，
  
  修回日期：2010-11-18，
  
  网络出版日期：2011-07-25，
  
  纸质出版日期：2011-07-25
- 稿件说明：
移动端阅览
杨永敏, 樊继壮, 赵杰. 基于超熵和模糊集理论的带钢表面缺陷分割[J]. 光学精密工程, 2011,19(7): 1651-1658

YANG Yong-min, FAN Ji-zhuang, ZHAO Jie. Steel strip surface defect segmentation based on excess entropy and fuzzy set theory[J]. Editorial Office of Optics and Precision Engineering, 2011,19(7): 1651-1658
杨永敏, 樊继壮, 赵杰. 基于超熵和模糊集理论的带钢表面缺陷分割[J]. 光学精密工程, 2011,19(7): 1651-1658 DOI： 10.3788/OPE.20111907.1651.

YANG Yong-min, FAN Ji-zhuang, ZHAO Jie. Steel strip surface defect segmentation based on excess entropy and fuzzy set theory[J]. Editorial Office of Optics and Precision Engineering, 2011,19(7): 1651-1658 DOI： 10.3788/OPE.20111907.1651.

摘要

由于冷轧带钢表面缺陷图像中存在过渡区

在图像分割过程中既要利用灰度信息也要利用空间结构信息才能取得好的分割效果。因此

本文研究了信息熵中的超熵以及模糊集理论

根据超熵可以用来测度图像的空间结构

模糊集可以描述出图像灰度过渡区的特性

提出了一种基于超熵和模糊集理论的图像分割算法。结合超熵和模糊集理论构建出模糊超熵

通过计算图像的最大模糊超熵所对应的最优隶属度函数参数组合确定了分割阈值

并利用该阈值完成图像分割。将该算法与Ostu以及一维最大模糊熵分割算法相比较

结果显示

本文算法能够准确地从背景中提取缺陷

有效地抑制了过分割现象。利用提出的误分割率和有效信息率对分割后的图像进行定量评价

结果表明

用本文算法分割后的图像有效信息率在3种方法中最高

均在82.7%以上

同时误分割率均低于2.1%。

Abstract

Because of the existence of transition zones in a cold rolling strip surface defect image

gray information and spatial structure information should be combined to segment images to obtain better image results.Therefore

the excess entropy of information entropy and fuzzy set theory were researched.As the excess entropy could be used to measure spatial structure of an image and the characteristic of image gray transition zone could be described well by the fuzzy set

an image threshold segmentation algorithm based on maximal fuzzy excess entropy was proposed.The fuzzy excess entropy was built by the combination of excess entropy and fuzzy set theory and the threshold was determined by the best membership function parameter combination according to the maximal fuzzy excess entropy value. Then

the image was segmented by using the threshold. Finally

the algorithm was compared with Ostu and 1D maximal fuzzy entropy segmentation algorithms. The experiment indicates that the proposed algorithm can extract the defect from a background exactly and can constrain the over-segmentation effectively. The quantificational evaluation of segmented image was performed by the wrong segmentation rate and effective information rate

and the results show that the effective information rate of the algorithm is higher than 82.7%

which is the maximal one among three methods.Meanwhile the wrong segmentation rate is below 2.1%.

关键词

Keywords

references

JUKKA I, JUHANI R, ARI V. Unsupervised segmentation of surface defects. Proceedings of the 13th International Conference on Pattern Recognition, Vienna, 1996, August, 25-29, 4:356-360.[2] 杨水山,何永辉,赵万生,等. Mean shift算法在带钢缺陷图像分割中的应用[J]. 西安电子科技大学学报,2007,34(6):1015-1018. YANG SH H, HE Y H, ZHAO W SH, et al. Application of the Mean shift algorithm in steel strip image segmentation[J]. Journal of Xidian University, 2007,34(6):1015-1018. (in Chinese)[3] SHARIFZADEH M,ALIREZAEE S,AMIRFATTAHI R, et al. Detection of steel defect using the image processing algorithms. Proceedings of the 12th IEEE International Multitopic Conference, Karachi, 2008, December , 23-24:125-127.[4] 李洪艳,曹建荣,谈文婷,等. 图像分割技术综述[J]. 山东建筑大学学报,2010,25(1):85-93. LI H Y, CAO J R, TAN W T, et al. Review of image segmentation technology[J]. Journal of Shang Dong Jian Zhu University, 2010,25(1):85-93.(in Chinese)[5] 程万胜,臧希喆 ,赵杰,等. 面向Ostu阈值搜索的PSO惯性因子改进方法[J]. 光学精密工程,2008,16(10):1907-1912. CHENG W S, ZANG X ZH, ZHAO J, et al. Modified strategy to inertia weight in PSO for searching threshold of Ostu rule[J]. Opt. Precision Eng., 2008,16(10):1907-1912.(in Chinese)[6] CHEN Y B,CHEN O T C. Image segmentation method using thresholds automatically determined from picture contents[J]. Journal on Image and Video Processing, 2009, 2009:1-15.[7] 张麒,汪源源,王威琪,等. 活动轮廓模型和Conturlet多分辨率分析分割血管内超声图像[J]. 光学精密工程,2008,16(11): 2303-2311. ZHANG Q, WANG Y Y, WANG W Q, et al. Intravascular ultrasound image segmentation based on active contour model and Contourlet multiresolution analysis[J]. Opt. Precision Eng., 2008, 16(11):2303-2311.(in Chinese)[8] MIHA M, DEJAN T, FRANJO P, et al. Real-time image segmentation for visual inspection of pharmaceutical tablets [J]. Machine Vision and Applications, 2009,22(1):145-156.[9] 张坤华,杨恒. 应用聚类和分形实现复杂背景下的扩展目标分割[J]. 光学精密工程,2009,17(7):1665-1671. ZHANG K H, YANG H. Segmentation for extended target in complex backgrounds based on clustering and fractal[J]. Opt. Precision Eng., 2009,17(7):1665-1671. (in Chinese)[10] ANIRBAN M, UJJWAL M. Unsupervised satellite image segmentation by combining SA based fuzzy clustering with support vector machine. Proceedings of the Seventh International Conference on Advances in Pattern Recognition, Kolkata, West Bengal, India, ICAPR, 2009:381-384.[11] 刘锁兰,杨静宇. 过渡区提取综述[J]. 中国工程科学,2007, 9(9):89-96. LIU S L, YANG J Y. Survey on extraction methods of transition region [J]. Engineering Sciences, 2007,9(9):89-96.(in Chinese)[12] BARDERA A,BOADA I,FEIXAS M, et al. Image segmentation using excess entropy [J]. Journal of Signal Processing Systems, 2009,54(1-3):205-214.[13] TORRENT A,BARDERA A, OLIVER A, et al. Breast density segmentation: a comparison o f clustering and region based techniques. Proceeding of the 9th International Workshop on Digital Mammography, IWDM 2008, Tucson, 2008, July, 20-23:9-16.[14] FELDMAN D P,CRUTCHFIELD J P. Structural information in two-dimensional patterns: entropy convergence and excess entropy [J]. Physical Review E, 2003, 67:051104/1-051104/9.[15] FELDMAN D. A brief introduction to: information theory, excess entropy and computational mechanics. Department of Physics, University of California, Berkeley, USA, 2002:13-22.

浏览量

399

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

融合条形注意力机制的电芯极片计算机断层图像分割

增强型金枪鱼群优化指数熵的砂粒显微图像分割

语义流引导采样结合注意力机制的脑肿瘤图像分割

混沌SSA优化多重熵阈值的骨料图像自动分割

基于高分辨率复合网络的皮肤病变分割