| Peer-Reviewed

Fine Crack Detection Algorithm Based on Improved SSD

Received: 25 May 2022     Accepted: 9 June 2022     Published: 16 June 2022
Views:       Downloads:
Abstract

The fine cracks on the workpiece surface are the significant potential hazard to safety of industrial production process. In order to identify fine cracks on the workpiece surface, an improved SSD (Single Shot MultiBox Detector) algorithm is built in this paper and applied to detect fine cracks. Based on the SSD network, the dilated convolution module is proposed in the convolutional operation to ensure access to global feature and by reducing the pooling layer treatment. In order to achieve the effective cracks detection, the cracks images are divided into two cases: obvious bold cracks and vague fine cracks, and mark them respectively. The obvious bold cracks are marked as "neg" and detected by SSD network framework, while the vague fine cracks are marked as "crack" and detected by SSD network with reduced pooling layer. This improvement is helpful to increase the detection accuracy of fine cracks. In this paper, the actual crack images are used to verify the improved algorithm. Results show that under the training and testing with workpiece crack data set, the improved algorithm can effectively detect fine cracks such that the detection precision toward the number of cracks in the image is higher than 80%. The aforementioned algorithms present potential application for the detection of fine cracks.

Published in International Journal on Data Science and Technology (Volume 8, Issue 2)
DOI 10.11648/j.ijdst.20220802.12
Page(s) 43-47
Creative Commons

This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.

Copyright

Copyright © The Author(s), 2022. Published by Science Publishing Group

Keywords

Fine Crack, SSD Network, Dilated Convolution, Crack Detection

References
[1] Hinton, G, E, et al. Reducing the Dimensionality of Data with Neural Networks. [J]. Science, 2006.
[2] Girshick R, Donahue J, Darrell T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation [J]. IEEE Computer Society, 2013.
[3] Girshick R. Fast R-CNN [J]. IEEE International Conference on Computer Vision (ICCV). Santiago: IEEE, 2015: 1440-1448.
[4] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks [J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39 (6): 1137-1149.
[5] He K, Zhang X, Ren S, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition [J]. Pattern Analysis & Machine Intelligence IEEE Transactions on, 2015, 37 (9): 1904-1916.
[6] RedmonJ, Divvala S, Girshick R, et al. You Only Look Once: Unified, Real-Time Object Detection [C]// Computer Vision & Pattern Recognition. IEEE, 2016.
[7] CUI Xiaoning, WANG Qicai, LI Sheng, et al. Intelligent recognition of cracks in double block sleeper based on YOLO-v5 [J]. Journal of the China Railway Society, 2022, 44 (4): 104-111.
[8] MA Jian, YAN Weidong, LIU Guoqi. Research on crack detection method of wooden ancient building based on YOLO v5 [J]. Journal of Shenyang Jianzhu University (Natural Science, 2021, 27 (5): 927-934.
[9] Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector [C]//European conference on computer vision. Springer, Cham, 2016: 21-37.
[10] Tang Cong, Liang Yongshun, Zheng Kedong, et al. Object detection method of multi-view SSD based on deep learning [J]. Infrared and Laser Engineering, 2018: 47 (1): 1-9.
[11] Xia Ye, Chen Limu, Wang Junjie, et al. Bridge active anti ship collision target detection method and application based on SSD [J]. Journal of Hunan University (NATURAL SCIENCE EDI TION), 2020, 47 (03): 97-105.
[12] Hongchen TAN, Jun ZHOU, Shengjing TIAN, et al. SFE-SSD: Shallow feature enhancement SSD for small object detection [J]. Journal of Mathematical Research with Applications, 2019, 39 (6): 733-744.
[13] Huang Jipeng, Shi Yinghuan, Gao Yang. Multiscale fastercnn detection algorithm for small targets [J]. Computer research and development, 2019, 56 (02): 319-327.
[14] Chen L C, Papandreou G, Kokkinos I, et al. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40 (4): 834-848.
[15] Li Maopeng. Research on target detection algorithm and pruning optimization based on improved SSD [D]. Nanjing: Nanjing University of Posts and telecommunications, 2020.
Cite This Article
  • APA Style

    Mai Ziying, Hu Shaolin, Huang Xiaomin, Ke Ye. (2022). Fine Crack Detection Algorithm Based on Improved SSD. International Journal on Data Science and Technology, 8(2), 43-47. https://doi.org/10.11648/j.ijdst.20220802.12

    Copy | Download

    ACS Style

    Mai Ziying; Hu Shaolin; Huang Xiaomin; Ke Ye. Fine Crack Detection Algorithm Based on Improved SSD. Int. J. Data Sci. Technol. 2022, 8(2), 43-47. doi: 10.11648/j.ijdst.20220802.12

    Copy | Download

    AMA Style

    Mai Ziying, Hu Shaolin, Huang Xiaomin, Ke Ye. Fine Crack Detection Algorithm Based on Improved SSD. Int J Data Sci Technol. 2022;8(2):43-47. doi: 10.11648/j.ijdst.20220802.12

    Copy | Download

  • @article{10.11648/j.ijdst.20220802.12,
      author = {Mai Ziying and Hu Shaolin and Huang Xiaomin and Ke Ye},
      title = {Fine Crack Detection Algorithm Based on Improved SSD},
      journal = {International Journal on Data Science and Technology},
      volume = {8},
      number = {2},
      pages = {43-47},
      doi = {10.11648/j.ijdst.20220802.12},
      url = {https://doi.org/10.11648/j.ijdst.20220802.12},
      eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijdst.20220802.12},
      abstract = {The fine cracks on the workpiece surface are the significant potential hazard to safety of industrial production process. In order to identify fine cracks on the workpiece surface, an improved SSD (Single Shot MultiBox Detector) algorithm is built in this paper and applied to detect fine cracks. Based on the SSD network, the dilated convolution module is proposed in the convolutional operation to ensure access to global feature and by reducing the pooling layer treatment. In order to achieve the effective cracks detection, the cracks images are divided into two cases: obvious bold cracks and vague fine cracks, and mark them respectively. The obvious bold cracks are marked as "neg" and detected by SSD network framework, while the vague fine cracks are marked as "crack" and detected by SSD network with reduced pooling layer. This improvement is helpful to increase the detection accuracy of fine cracks. In this paper, the actual crack images are used to verify the improved algorithm. Results show that under the training and testing with workpiece crack data set, the improved algorithm can effectively detect fine cracks such that the detection precision toward the number of cracks in the image is higher than 80%. The aforementioned algorithms present potential application for the detection of fine cracks.},
     year = {2022}
    }
    

    Copy | Download

  • TY  - JOUR
    T1  - Fine Crack Detection Algorithm Based on Improved SSD
    AU  - Mai Ziying
    AU  - Hu Shaolin
    AU  - Huang Xiaomin
    AU  - Ke Ye
    Y1  - 2022/06/16
    PY  - 2022
    N1  - https://doi.org/10.11648/j.ijdst.20220802.12
    DO  - 10.11648/j.ijdst.20220802.12
    T2  - International Journal on Data Science and Technology
    JF  - International Journal on Data Science and Technology
    JO  - International Journal on Data Science and Technology
    SP  - 43
    EP  - 47
    PB  - Science Publishing Group
    SN  - 2472-2235
    UR  - https://doi.org/10.11648/j.ijdst.20220802.12
    AB  - The fine cracks on the workpiece surface are the significant potential hazard to safety of industrial production process. In order to identify fine cracks on the workpiece surface, an improved SSD (Single Shot MultiBox Detector) algorithm is built in this paper and applied to detect fine cracks. Based on the SSD network, the dilated convolution module is proposed in the convolutional operation to ensure access to global feature and by reducing the pooling layer treatment. In order to achieve the effective cracks detection, the cracks images are divided into two cases: obvious bold cracks and vague fine cracks, and mark them respectively. The obvious bold cracks are marked as "neg" and detected by SSD network framework, while the vague fine cracks are marked as "crack" and detected by SSD network with reduced pooling layer. This improvement is helpful to increase the detection accuracy of fine cracks. In this paper, the actual crack images are used to verify the improved algorithm. Results show that under the training and testing with workpiece crack data set, the improved algorithm can effectively detect fine cracks such that the detection precision toward the number of cracks in the image is higher than 80%. The aforementioned algorithms present potential application for the detection of fine cracks.
    VL  - 8
    IS  - 2
    ER  - 

    Copy | Download

Author Information
  • School of Automation, Guangdong University of Petrochemical Technology, Maoming, China

  • School of Automation, Guangdong University of Petrochemical Technology, Maoming, China

  • School of Automation and Information Engineering, Xi’an University of Technology, Xi’an, China

  • School of Automation, Guangdong University of Petrochemical Technology, Maoming, China

  • Sections