Please wait a minute...
浙江大学学报(工学版)  2020, Vol. 54 Issue (3): 427-434    DOI: 10.3785/j.issn.1008-973X.2020.03.001
机械工程     
基于特征与形貌重构的轴件表面缺陷检测方法
冯毅雄(),李康杰,高一聪*(),郑浩,谭建荣
浙江大学 流体动力与机电系统国家重点实验室, 浙江 杭州 310027
Shaft surface defect detection method based on feature and morphology reconstruction
Yi-xiong FENG(),Kang-jie LI,Yi-cong GAO*(),Hao ZHEN,Jian-rong TAN
State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University, Hangzhou 310027, China
 全文: PDF(3598 KB)   HTML
摘要:

针对轴件表面缺陷机器视觉检测方法中的水渍残留误检率高和人工复检效率低问题,提出一种基于特征与形貌重构的轴件表面缺陷检测方法. 对轴件工业高速线扫描图像进行预处理,基于改进的阀值迭代算法完成图像分割,通过去除背景、噪点和干扰提取缺陷图像. 建立基于曲线簇包络轮廓的轴件表面缺陷特征模型,结合分割图像各连通域的面积、面积占比、粗短度训练逻辑回归分类器,对凹坑、裂纹和麻点等轴件表面典型缺陷进行识别,并结合图像深度信息进行缺陷形貌重构,消除水渍等伪缺陷,提高轴件表面缺陷检测鲁棒性. 实验结果表明,所提出的轴件表面缺陷检测方法有效,具有较高的缺陷识别率和鲁棒性能,平均识别时间为3.69 s,缺陷轴加权识别率为98.86%,可以对3类典型缺陷和水渍进行准确识别.

关键词: 轴件表面缺陷缺陷特征提取形貌重构缺陷分类    
Abstract:

A surface defect detection method based on feature and morphology reconstruction was proposed, aiming at the problems of high misdetection rate of water stain residue and low efficiency of manual re-inspection in machine vision detection of defects in shaft parts surface. The high-speed industry line scanning image of the shaft part was preprocessed, and image segmentation was completed based on the improved threshold iteration algorithm. Defect image was extracted by removing background, noise, and interference. A surface defect feature model of shaft parts based on envelope contour of curve cluster was established. Combining the area, the proportion of area and coarseness of each connected area in the segmented image, a logistic regression classifier was trained to recognize the typical surface defects of pits, cracks, and pits of shaft parts. Combining with image depth information, defect morphology reconstruction was carried out to eliminate pseudo-defects such as water stain, so as to improve the robustness of surface defect detection of shaft parts. The experimental results show that this method is effective for surface defect detection of shaft parts with high defect recognition rate and robustness. The proposed method has an average recognition time of 3.69 seconds and a weighted recognition rate of 98.86%, which can accurately identify three kinds of typical defects and pseudo-defects.

Key words: defect of shaft surface    defect feature extraction    shape reconstruction    defect classification
收稿日期: 2019-05-05 出版日期: 2020-03-05
CLC:  TN 911.73  
基金资助: 国家重点研发计划资助项目(2017YFB1301201);国家自然科学基金资助项目(51675477,51775489,51805472);浙江省自然科学基金资助项目(LZ18E050001)
通讯作者: 高一聪     E-mail: fyxtv@zju.edu.cn;gaoyicong@zju.edu.cn
作者简介: 冯毅雄(1975—),男,教授,博士,从事现代机械设计理论与方法研究. orcid.org/0000-0001-7397-2482. E-mail: fyxtv@zju.edu.cn
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  
冯毅雄
李康杰
高一聪
郑浩
谭建荣

引用本文:

冯毅雄,李康杰,高一聪,郑浩,谭建荣. 基于特征与形貌重构的轴件表面缺陷检测方法[J]. 浙江大学学报(工学版), 2020, 54(3): 427-434.

Yi-xiong FENG,Kang-jie LI,Yi-cong GAO,Hao ZHEN,Jian-rong TAN. Shaft surface defect detection method based on feature and morphology reconstruction. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 427-434.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2020.03.001        http://www.zjujournals.com/eng/CN/Y2020/V54/I3/427

图 1  轴件扫描图像获取系统硬件结构示意图
图 2  轴件线扫描图像
图 3  线扫描图像中沿A1、A2、A3直线灰度分布
图 4  轴件图像归一化灰度直方图
图 5  轴件图像迭代阈值算法处理结果
图 6  轴件二值图竖直方向累加值
图 7  轴件表面缺陷模型
图 8  实际缺陷图像与对应缺陷模型
图 9  轴件表面缺陷特征示意图
图 10  伪缺陷水渍图像
图 11  轴件表面4类缺陷的三维重建图
图 12  缺陷识别算法整体流程图
图 13  合格轴检测结果
图 14  缺陷轴检测结果
缺陷识别统计项 数值 缺陷识别统计项 数值
TP 955 TN 21
FN 22 t/s 3.69
FP 1 F1/% 98.86
表 1  缺陷轴识别结果
类别 n1 n2 n3 n4 η
训练集 25 25 25 25 82
测试集 5 5 5 5 75
表 2  分类器实验结果
1 谢中灏. 重载轴毂复合连接扭矩传递特性的研究[D]. 杭州: 浙江大学, 2014: 1-2.
XIE Zhong-hao. Torque transfer characteristics of heavy-duty hub-axle composite connections[D]. Hangzhou: Zhejiang University, 2014: 1-2.
2 高义林. 基于线阵CCD的车轮表面荧光磁粉探伤系统的研究[D]. 合肥: 安徽大学, 2010: 10-13.
GAO Yi-lin. Research on the fluorescent magnetic particle detection system for wheel surface based on linear array CCD[D]. Hefei: Anhui University, 2010: 10-13.
3 BHUTADA G G, ANAND R S, SAXENA S C Edge preserved image enhancement using adaptive fusion of images denoised by wavelet and curvelet transform[J]. Digital Signal Processing, 2011, 21 (1): 118- 130
doi: 10.1016/j.dsp.2010.09.002
4 IBRAHIM H, ABDALAMEER AK Improvement of quantized adaptive switching median filter for impulse noise reduction in gray-scale digital images[J]. Journal Citation Reports, 2019, 27 (1): 580- 594
5 MUKHERJEE D, MUKHOPADHYAY S, BISWAS GP FPGA based parallel implementation of morphological operators for 2d gray-level images[J]. Arabian Journal for Science and Engineering, 2017, 42 (8): 3191- 3206
doi: 10.1007/s13369-017-2429-y
6 ZHANG W, LI W S, YAN J W, et al Adaptive threshold selection for background removal in fringe projection profilometry[J]. Optics and Lasers in Engineering, 2016, 90: 209- 216
7 DEVI KD, MAHESWARI P U Digital acquisition and character extraction from stone inscription images using modified fuzzy entropy-based adaptive thresholding[J]. Soft Computing, 2019, 23 (8): 2611- 2626
doi: 10.1007/s00500-018-3610-2
8 YUE J, WANG Z X, HAN J, et al A target extraction method of infrared image based on edge and transition region[J]. Spectroscopy and Spectral Analysis, 2018, 38 (6): 1729- 1735
9 YUN J P, KIM D, KIM K Vision-based surface defect inspection for thick steel plates[J]. Optical Engineering, 2017, 56 (5): 1- 13
10 GAN J R, LI Q T, WANG J Z A hierarchical extractor-based visual rail surface inspection system[J]. IEEE Sensors Journal, 2017, 17 (23): 7935- 7944
doi: 10.1109/JSEN.2017.2761858
11 MENTOURI Z, MOUSSAOUI A, BOUDJEHEM D, et al Steel strip surface defect identification based on binarized statistical features[J]. University Politehnica of Bucharest Scientific Bulletin Series B-Chemistry and Materials Science, 2018, 80 (4): 145- 156
12 SAYED M S. Robust fabric defect detection algorithm using entropy filtering and minimum error thresholding[C] // IEEE International Midwest Symposium on Circuits and Systems. Abu Dhabi: IEEE, 2017: 1-4.
13 LI M X, LIU Q X, HE C Y Research and achievement on cigarette label printing defect detection algorithm[J]. Applied Mechanics and Materials, 2012, 200: 689- 693
doi: 10.4028/www.scientific.net/AMM.200.689
14 ZHANG Y, LEFEBVRE D, LI Q Automatic detection of defects in tire radiographic images[J]. IEEE Transactions on Automation Science and Engineering, 2017, 14 (3): 1378- 1386
doi: 10.1109/TASE.2015.2469594
15 何再兴, 张树有, 黄长林 基于缺陷特征与种子填充的轮毂缺陷检测方法[J]. 浙江大学学报: 工学版, 2009, 43 (7): 1230- 1237
HE Zai-xing, ZHANG Shu-you, HUANG Chang-lin Hub defect detection method based on defect characteristics and seed filling[J]. Journal of Zhejiang University: Engineering Science, 2009, 43 (7): 1230- 1237
16 王宣银, 梁冬泰 基于多元图像分析的表面缺陷检测算法[J]. 浙江大学学报: 工学版, 2010, 44 (3): 448- 452
WANG Xuan-yin, LIANG Dong-tai Surface defect detection algorithm based on multivariate image analysis[J]. Journal of Zhejiang University: Engineering Science, 2010, 44 (3): 448- 452
17 李漫丽, 赵鹏 基于图像融合的木板表面缺陷检测研究[J]. 液晶与显示, 2016, 31 (9): 882- 888
LI Man-li, ZHAO Peng Image fusion based wood surface defect detection[J]. LCD and Display, 2016, 31 (9): 882- 888
18 LU J L, LIN M X An online surface defects detection system for step-axis[J]. Advanced Materials Research, 2012, 472: 954- 957
19 孙雪晨, 姜肖楠, 傅瑶, 等 基于机器视觉的凸轮轴表面缺陷检测系统[J]. 红外与激光工程, 2013, 42 (6): 1647- 1653
SUN Xue-chen, JIANG Xiao-nan, FU Yao, et al Camshaft surface defect detection system based on machine vision[J]. Infrared and Laser Engineering, 2013, 42 (6): 1647- 1653
doi: 10.3969/j.issn.1007-2276.2013.06.047
20 SHAO L, MU Y, GUO P, et al Research of image segmentation based on iterative threshold[J]. Applied Mechanics and Materials, 2015, 701-702: 330- 333
21 TSAI P S, SHAH M Shape from shading using linear approximation[J]. Image and Vision Computing, 1995, 12 (8): 487- 498
[1] 贾真,董文德,徐贵力,朱士鹏. 基于马尔科夫专家场的泊松噪声图像去噪方法[J]. 浙江大学学报(工学版), 2020, 54(6): 1164-1169.
[2] 王志, 朱世强, 卜琰, 郭振民. 改进导向滤波器立体匹配算法[J]. 浙江大学学报(工学版), 2016, 50(12): 2262-2269.
[3] 朱株,刘济林. 基于马尔科夫随机场的三维激光雷达路面实时分割[J]. 浙江大学学报(工学版), 2015, 49(3): 464-469.
[4] 吴一全,张晓杰,吴诗婳,张生伟. 基于混沌PSO或分解的二维最小误差阈值分割[J]. J4, 2011, 45(7): 1198-1205.
[5] 毛锋, 张树有, 施岳定. 多分辨轮廓匹配的成型制件外形检测[J]. J4, 2011, 45(6): 1013-1020.
[6] 葛鹏,李奇, 冯华君,徐之海,陈跃庭. 双三次样条插值联合变换相关器亚像素探测技术[J]. J4, 2010, 44(11): 2198-2202.
[7] 谢强军, 侯迪波, 黄平捷, 张光新, 周泽魁. 基于半隐差分的单参数水平集快速分割[J]. J4, 2010, 44(8): 1496-1501.
[8] 毛锋, 张树有, 黄长林. 图像散布图和小波多分辨分析的模具异物检测[J]. J4, 2009, 43(10): 1749-1756.
[9] 凌波, 顾伟康, 杜歆. H264整帧丢失下的错误隐藏机制[J]. J4, 2009, 43(09): 1732-1738.