Improvement and application of YOLOv3 for defect detection of smart phone glass covers

Ji-gang WU; Yuan CHENG; Jun SHAO; De-qiang YANG

doi:10.37188/CJLCD.2021-0172

您当前的位置：

首页 >

文章列表页 >

Improvement and application of YOLOv3 for defect detection of smart phone glass covers

Image Processing | 更新时间：2021-12-10

- Improvement and application of YOLOv3 for defect detection of smart phone glass covers
- Chinese Journal of Liquid Crystals and Displays Vol. 36, Issue 12, Pages: 1728-1736(2021)
- 作者机构：
  
  湖南科技大学机械设备健康维护湖南省重点实验室, 湖南湘潭 411201
- 作者简介：
- 基金信息：
  
  Supported by National Natural Science Foundation of China(51775181)
- DOI：10.37188/CJLCD.2021-0172
  CLC： TP391.41
- Received：30 June 2021，
  
  Revised：03 August 2021，
  
  Published：2021-12
- 稿件说明：
移动端阅览
Ji-gang WU, Yuan CHENG, Jun SHAO, et al. Improvement and application of YOLOv3 for defect detection of smart phone glass covers[J]. Chinese journal of liquid crystals and displays, 2021, 36(12): 1728-1736.
DOI：

Ji-gang WU, Yuan CHENG, Jun SHAO, et al. Improvement and application of YOLOv3 for defect detection of smart phone glass covers[J]. Chinese journal of liquid crystals and displays, 2021, 36(12): 1728-1736. DOI： 10.37188/CJLCD.2021-0172.

摘要

针对智能手机玻璃盖板缺陷检测方法存在检测柔性差、良率低、检测时间长等问题，提出一种改进YOLOv3的智能手机玻璃盖板缺陷检测方法。在特征提取网络方面增加通道注意力机制以解决缺陷特征不明显的问题，在特征检测网络方面增加了104×104维度大小的特征图以解决缺陷多尺度的问题，最后对模型进行剪枝减少模型参数，提高缺陷检测速度。从智能手机玻璃盖板生产现场获得涵盖崩边、坑点、脏污和划痕等4类缺陷的图片构建缺陷数据集，对本文提出的方法和Faster R-CNN、YOLOv3、YOLOv4等算法进行对比实验和分析。实验结果表明，本文提出方法的检测平均精度均值（mean Average Precision，mAP）为81.0%，检测速度为43.1 fps。相比原始YOLOv3算法，检测mAP提升了3%，检测速度增加了6.7 fps，相比于其他深度学习算法，检测速度和检测精度均有所提升。所提方法满足智能手机玻璃盖板工业生产现场缺陷高精度、高效检测的需要。

Abstract

To address the problems of poor detection flexibility

low yield rate and long detection time of smartphone glass cover defect detection methods

an improved YOLOv3 defect detection method for smartphone glass cover is proposed. A channel attention mechanism is added to the feature extraction network to solve the problem of inconspicuous defect features

a feature map of 104×104 dimensional size is added to the feature detection network to solve the problem of multi-scale defects

and finally the model is pruned to reduce the model parameters to improve the defect detection speed. The defect dataset is constructed by obtaining images covering four types of defects

such as chipped edge

pit

dirty and scratches

from the production site of smartphone glass cover. The proposed method and algorithms such as Faster R-CNN

YOLOv3 and YOLOv4 are compared for experiments and analysis. The experimental results show that the detection mAP (mean average precision) of the proposed method is 81.0% and the detection speed is 43.1 fps. Compared with the original YOLOv3 algorithm

the detection mAP is improved by 3% and the detection speed is increased by 6.7 fps. Compared with other deep learning algorithms

the detection speed and detection precision are improved. The proposed method meets the need for high-precision and efficient detection of defects in the industrial production site of smartphone glass covers.

关键词

Keywords

references

李晚秋 , 乔翠霞 . 基于SCP范式的中国智能手机产业分析 . 产业经济评论 , 2018 . ( 6 ): 51 - 62 . https://doc.paperpass.com/journal/20131785qnkx-jsb.html https://doc.paperpass.com/journal/20131785qnkx-jsb.html .

W Q LI , C X QIAO . Analysis of China's smartphone industry based on SCP paradigm . Review of Industrial Economics , 2018 . ( 6 ): 51 - 62 . https://doc.paperpass.com/journal/20131785qnkx-jsb.html https://doc.paperpass.com/journal/20131785qnkx-jsb.html .

简川霞 , 高健 . 手机玻璃屏表面缺陷视觉检测方法研究 . 包装工程 , 2018 . 39 ( 5 ): 16 - 21 . DOI: 10.19554/j.cnki.1001-3563.2018.05.004 http://doi.org/10.19554/j.cnki.1001-3563.2018.05.004 .

C X JIAN , J GAO . Visual detection method for surface defect of mobile phone screen glass . Packaging Engineering , 2018 . 39 ( 5 ): 16 - 21 . DOI: 10.19554/j.cnki.1001-3563.2018.05.004 http://doi.org/10.19554/j.cnki.1001-3563.2018.05.004 .

简川霞. 手机玻璃屏表面缺陷机器视觉检测与分类方法研究[D]. 广州: 广东工业大学, 2017. https://d.wanfangdata.com.cn/thesis/Y3345758

JIAN C X. Surface defect detection and classification of mobile phone screen glass based on machine vision[D]. Guangzhou: Guangdong University of Technology, 2017. (in Chinese)

C X JIAN , J GAO , Y H AO . Automatic surface defect detection for mobile phone screen glass based on machine vision . Applied Soft Computing , 2017 . 52 348 - 358 . DOI: 10.1016/j.asoc.2016.10.030 http://doi.org/10.1016/j.asoc.2016.10.030 .

L Q LIANG , D LI , X FU , 等 . Touch screen defect inspection based on sparse representation in low resolution images . Multimedia Tools and Applications , 2016 . 75 ( 5 ): 2655 - 2666 . DOI: 10.1007/s11042-015-2559-8 http://doi.org/10.1007/s11042-015-2559-8 .

张建国 , 李颖 , 齐家坤 , 等 . 基于机器视觉的手机屏幕表面划痕检测研究 . 应用光学 , 2020 . 41 ( 5 ): 984 - 989 . DOI: 10.5768/JAO202041.0502008 http://doi.org/10.5768/JAO202041.0502008 .

J G ZHANG , Y LI , J K QI , 等 . Surface scratch detection of mobile phone screen based on machine vision . Journal of Applied Optics , 2020 . 41 ( 5 ): 984 - 989 . DOI: 10.5768/JAO202041.0502008 http://doi.org/10.5768/JAO202041.0502008 .

郭波 , 管菊花 , 黄志开 . 基于改进Otsu算法的TFT-LCD点缺陷自动光学检测系统 . 液晶与显示 , 2018 . 33 ( 3 ): 221 - 227 . DOI: 10.3788/YJYXS20183303.0221 http://doi.org/10.3788/YJYXS20183303.0221 .

B GUO , J H GUAN , Z K HUANG . Automatic optical detection system for TFT-LCD spot-type defect based on improved Otsu algorithm . Chinese Journal of Liquid Crystals and Displays , 2018 . 33 ( 3 ): 221 - 227 . DOI: 10.3788/YJYXS20183303.0221 http://doi.org/10.3788/YJYXS20183303.0221 .

李娜 , 姜志 , 王军 , 等 . 基于Faster R-CNN的仪表识别方法 . 液晶与显示 , 2020 . 35 ( 12 ): 1291 - 1298 . DOI: 10.37188/YJYXS20203512.1291 http://doi.org/10.37188/YJYXS20203512.1291 .

N LI , Z JIANG , J WANG , 等 . Instrument recognition method based on Faster R-CNN . Chinese Journal of Liquid Crystals and Displays , 2020 . 35 ( 12 ): 1291 - 1298 . DOI: 10.37188/YJYXS20203512.1291 http://doi.org/10.37188/YJYXS20203512.1291 .

范丽丽 , 赵宏伟 , 赵浩宇 , 等 . 基于深度卷积神经网络的目标检测研究综述 . 光学精密工程 , 2020 . 28 ( 5 ): 1152 - 1164 . http://qikan.cqvip.com/Qikan/Article/Detail?id=7101529202 http://qikan.cqvip.com/Qikan/Article/Detail?id=7101529202 .

L L FAN , H W ZHAO , H Y ZHAO , 等 . Survey of target detection based on deep convolutional neural networks . Optics and Precision Engineering , 2020 . 28 ( 5 ): 1152 - 1164 . http://qikan.cqvip.com/Qikan/Article/Detail?id=7101529202 http://qikan.cqvip.com/Qikan/Article/Detail?id=7101529202 .

REDMON J, DIVVALA S, GIRSHICK R, et al . You only look once: unified, real-time object detection[C]// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition . Las Vegas: IEEE, 2016: 779-788.

余永维 , 韩鑫 , 杜柳青 . 基于Inception-SSD算法的零件识别 . 光学精密工程 , 2020 . 28 ( 8 ): 1799 - 1809 . DOI: 10.3788/OPE.20202808.1799 http://doi.org/10.3788/OPE.20202808.1799 .

Y W YU , X HAN , L Q DU . Target part recognition based inception-SSD algorithm . Optics and Precision Engineering , 2020 . 28 ( 8 ): 1799 - 1809 . DOI: 10.3788/OPE.20202808.1799 http://doi.org/10.3788/OPE.20202808.1799 .

李红光 , 于若男 , 丁文锐 . 基于深度学习的小目标检测研究进展 . 航空学报 , 2021 . 42 ( 7 ): 100 - 118 . http://cpfd.cnki.com.cn/Article/CPFDTOTAL-XFWF201911001036.htm http://cpfd.cnki.com.cn/Article/CPFDTOTAL-XFWF201911001036.htm .

H G LI , R N YU , W R DING . Research development of small object traching based on deep learning . Acta Aeronautica et Astronautica Sinica , 2021 . 42 ( 7 ): 100 - 118 . http://cpfd.cnki.com.cn/Article/CPFDTOTAL-XFWF201911001036.htm http://cpfd.cnki.com.cn/Article/CPFDTOTAL-XFWF201911001036.htm .

鲁越. 基于深度学习的手机玻璃缺陷分类检测[D]. 郑州: 郑州大学, 2019. https://d.wanfangdata.com.cn/thesis/ChJUaGVzaXNOZXdTMjAyMDEwMjgSCUQwMTY5OTg3MhoIeGlpeGVyeXc%3D

LU Y. Classification and detection of mobile phone glass defects based on deep learning[D]. Zhengzhou: Zhengzhou University, 2019. (in Chinese).

欧阳健燊. 基于深度学习的手机屏幕缺陷检测研究[D]. 广州: 华南理工大学, 2020. https://d.wanfangdata.com.cn/thesis/D01761264

OUYANG J S. Research on mobile phone screen defect detection method based on deep learning[D]. Guangzhou: South China University of Technology, 2020. (in Chinese).

REDMON J, FARHADI A. YOLOv3: an incremental improvement[J]. arXiv: 1804.02767, 2018.

巫明秀 , 吴谨 , 张晨 , 等 . 基于改进YOLOv3的棉花异性纤维检测 . 液晶与显示 , 2020 . 35 ( 11 ): 1195 - 1203 . DOI: 10.37188/YJYXS20203511.1195 http://doi.org/10.37188/YJYXS20203511.1195 .

M X WU , J WU , C ZHANG , 等 . Detection of foreign fiber in cotton based on improved YOLOv3 . Chinese Journal of Liquid Crystals and Displays , 2020 . 35 ( 11 ): 1195 - 1203 . DOI: 10.37188/YJYXS20203511.1195 http://doi.org/10.37188/YJYXS20203511.1195 .

李维刚 , 叶欣 , 赵云涛 , 等 . 基于改进YOLOv3算法的带钢表面缺陷检测 . 电子学报 , 2020 . 48 ( 7 ): 1284 - 1292 . DOI: 10.3969/j.issn.0372-2112.2020.07.006 http://doi.org/10.3969/j.issn.0372-2112.2020.07.006 .

W G LI , X YE , Y T ZHAO , 等 . Strip steel surface defect detection based on improved YOLOv3 algorithm . Acta Electronica Sinica , 2020 . 48 ( 7 ): 1284 - 1292 . DOI: 10.3969/j.issn.0372-2112.2020.07.006 http://doi.org/10.3969/j.issn.0372-2112.2020.07.006 .

陈宏彩 , 任亚恒 , 郝存明 , 等 . 基于YOLOv3的医药玻璃瓶缺陷检测方法 . 包装工程 , 2020 . 41 ( 7 ): 241 - 246 . http://www.packjour.cn/bzgcgk/ch/reader/view_abstract.aspx?file_no=202007034&flag=1 http://www.packjour.cn/bzgcgk/ch/reader/view_abstract.aspx?file_no=202007034&flag=1 .

H C CHEN , Y H REN , C M HAO , 等 . Defect detection method for medical glass bottles based on YOLOv3 . Packaging Engineering , 2020 . 41 ( 7 ): 241 - 246 . http://www.packjour.cn/bzgcgk/ch/reader/view_abstract.aspx?file_no=202007034&flag=1 http://www.packjour.cn/bzgcgk/ch/reader/view_abstract.aspx?file_no=202007034&flag=1 .

HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]// Proceedings of 2018 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City: IEEE, 2018: 7132-7141.

张慧 , 王坤峰 , 王飞跃 . 深度学习在目标视觉检测中的应用进展与展望 . 自动化学报 , 2017 . 43 ( 8 ): 1289 - 1305 . DOI: 10.16383/j.aas.2017.c160822 http://doi.org/10.16383/j.aas.2017.c160822 .

H ZHANG , K F WANG , F Y WANG . Advances and perspectives on applications of deep learning in visual object detection . Acta Automatica Sinica , 2017 . 43 ( 8 ): 1289 - 1305 . DOI: 10.16383/j.aas.2017.c160822 http://doi.org/10.16383/j.aas.2017.c160822 .

LIU Z, LI J G, SHEN Z Q, et al . Learning efficient convolutional networks through network slimming[C]// Proceedings of 2017 IEEE International Conference on Computer Vision . Venice: IEEE, 2017: 2755-2763.

Views

261

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Defect detection of small object solder joints based on improved YOLOv7

DFNet： efficient decoder-free semantic segmentation networks

Detection of reflective surface defects based on phase measuring deflectometry

TFT array defect detection based on AttentionGAN and morphological reconstruction

Related Author

De-qiang YANG

Jun SHAO

Yuan CHENG

Ji-gang WU

GAO Junwei

CAO Wei

LIU Zhaolong

HAN Jun

Related Institution

Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment， Hunan University of Science and Technology

Shandong Key Laboratory of Industrial Control Technology

Qingdao International Airport Group Co. Ltd.

School of Automation， Qingdao University

College of Electrical Engineering and Automation， Xiamen University of Technology

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰