KIPS Transactions on Software and Data Engineering (정보처리학회논문지:소프트웨어 및 데이터공학)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Tire Surface Defect Detection Method Using Depth Image

깊이 이미지를 이용한 타이어 표면 결함 검출 방법에 관한 연구

  • 김현석 (한국산업기술대학교 스마트팩토리융합과) ;
  • 고동범 (한국전자통신연구원 차세대시스템SW) ;
  • 이원곡 (한국산업기술대학교 인공지능기술사업화연구소) ;
  • 배유석 (한국산업기술대학교 컴퓨터공학과)
  • Received : 2021.10.01
  • Accepted : 2021.12.06
  • Published : 2022.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, research on smart factories triggered by the 4th industrial revolution is being actively conducted. Accordingly, the manufacturing industry is conducting various studies to improve productivity and quality based on deep learning technology with robust performance. This paper is a study on the method of detecting tire surface defects in the visual inspection stage of the tire manufacturing process, and introduces a tire surface defect detection method using a depth image acquired through a 3D camera. The tire surface depth image dealt with in this study has the problem of low contrast caused by the shallow depth of the tire surface and the difference in the reference depth value due to the data acquisition environment. And due to the nature of the manufacturing industry, algorithms with performance that can be processed in real time along with detection performance is required. Therefore, in this paper, we studied a method to normalize the depth image through relatively simple methods so that the tire surface defect detection algorithm does not consist of a complex algorithm pipeline. and conducted a comparative experiment between the general normalization method and the normalization method suggested in this paper using YOLO V3, which could satisfy both detection performance and speed. As a result of the experiment, it is confirmed that the normalization method proposed in this paper improved performance by about 7% based on mAP 0.5, and the method proposed in this paper is effective.

최근 4차 산업혁명으로 촉발된 스마트공장에 관한 연구가 활발히 진행되고 있다. 이에 따라 제조업에서는 강건한 성능의 딥러닝 기술을 바탕으로 생산성 향상과 품질 향상을 위해 다양한 연구를 진행 중이다. 본 논문은 타이어 제조공정의 육안검사 단계에서 타이어 표면 결함을 검출하는 방법에 관한 연구로서 3D 카메라를 통해 취득한 깊이 이미지를 이용한 타이어 표면 결함 검출 방법을 소개한다. 본 연구에서 다루는 타이어 표면 깊이 이미지는 타이어 표면의 얕은 깊이로 인해 발생되는 낮은 깊이 대비와 데이터 취득 환경으로 인해 기준 깊이 값의 차이가 발생하는 문제가 있다. 그리고 제조업의 특성상 검출 성능과 함께 실시간으로 처리될 수 있는 성능을 지닌 알고리즘이 요구된다. 따라서, 본 논문에서는 타이어 표면 결함 검출 알고리즘이 복잡한 알고리즘 파이프라인으로 구성되지 않도록 상대적으로 단순한 방법들을 통해 깊이 이미지를 정규화하는 방법을 연구하였으며 검출 성능과 속도를 모두 만족할 수 있는 딥러닝 방법인 YOLO V3를 이용하여 일반적인 정규화 방법과 본 논문에서 제안하는 정규화 방법의 비교 실험을 진행하였다. 실험의 결과로 본 논문에서 제안한 정규화 방법으로 mAP 0.5 기준 약 7% 성능이 향상된 것을 확인하였으며 본 논문에서 제시한 방법이 효과적임을 보였다.

Keywords

Acknowledgement

본 연구는 과학기술정보통신부 및 정보통신기획평가원의 Grand ICT연구센터지원사업의 연구결과로 수행되었음(IITP-2020-0-101741).

References

  1. Ministry of Culture, Sports and Tourism, Smart Factory (Intelligent Factory) [Internet], https://www.korea.kr/special/policyCurationView.do?newsId=148866604.
  2. Y. J. Cho, "The strategy for Smart Factory of Korea in the era of the Industry 4.0," Communications of the Korean Institute of Information Scientists and Engineers, Vol.35, No.6, pp.40-48, Jun. 2017.
  3. Automation Technology, 3D Laser Sensors for Measurements by Means of Laser-Triangulation [Internet], https://www.automationtechnology.de/cms/en/3d-laser-sensors-for-measurements-by-means-of-laser-triangulation.
  4. H. J. Park, Y. W. Lee and B. G. Kim, "Efficient tire wear and defect detection algorithm based on deep learnin," Journal of Korea Multimedia Society, Vol.24, No.8, pp.1026-1034, Aug. 2021. https://doi.org/10.9717/KMMS.2021.24.8.1026
  5. K. He, G. Gkioxari, P. Dollar, and R. Girshick, "Mask R-CNN," in Proceedings of the IEEE International Conference on Computer Vision, pp.2961-2969, 2017.
  6. R. Wang, Q. Guo, S. Lu, and C. Zhang, "Tire defect detection using fully convolutional network," IEEE Access, Vol.7, pp.43502-43510, Jan. 2019. https://doi.org/10.1109/access.2019.2908483
  7. P. Wang, P. Chen, Y. Yuan, D. Liu, Z. Huang, X. Hou, and G. Cottrell, "Understanding convolution for semantic segmentation," in Proceeding of the IEEE Winter Conference on Applications of Computer Vision, Jun. 2018.
  8. J. Long, E. Shelhamer, and T. Darrell, "Fully convolutional networks for semantic segmentation," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, Mar. 2015.
  9. H. Y. Chan, F. K. Lam, and H. Zhu, "Adaptive thresholding by variational method," IEEE Transactions on Image Processing, Vol.7, No.3, Mar. 1998.
  10. P. Roy, G. Dey, S. Dutta, S. Chakraborty, N. Dey, and R. Ray, "Adaptive thresholding: A comparative study," in Proceeding of the International Conference on Control, Instrumentation, Communication and Computational Technologies, Jul. 2014.
  11. I. S. Oh, "Binary image," in Computer Vision, 3rd ed, Hanbit Academy.Inc, pp.68-71, 2018.
  12. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You only look once: Unified, real-time object detection," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.779-788, 2016.
  13. J. Redmon and A. Farhadi "YOLO9000: Better, faster, stronger," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.7263-7271, 2017.
  14. J. Redmon and A. Farhadi, "Yolov3: An incremental improvement," arXiv preprint arXiv:1804.02767, Apr. 2018.
  15. Y. H. Lee and Y. S. Kim, "Comparison of CNN and YOLO for object detection," Journal of the Semiconductor & Display Technology, Vol.19, No.1. Mar. 2020.
  16. N. S. Punn, S. K. Sonbhadra, S. Agarwal, and G. Rai, "Monitoring COVID-19 social distancing with person detection and tracking via fine-tuned YOLO v3 and Deepsort techniques," arXiv preprint arXiv:2005.01385, Apr. 2021.