DOI QR코드

DOI QR Code

센서 및 영상데이터를 이용한 용접 비드 불량검사 소프트웨어 구현

Software Implementation of Welding Bead Defect Detection using Sensor and Image Data

  • 이재은 (부경대학교 IT융합응용공학과) ;
  • 김영봉 (부경대학교 IT융합응용공학과) ;
  • 김종남 (부경대학교 IT융합응용공학과)
  • Lee, Jae Eun (Dept. of IT Convergence & Applications Engineering, Pukyong National University) ;
  • Kim, Young-Bong (Dept. of IT Convergence & Applications Engineering, Pukyong National University) ;
  • Kim, Jong-Nam (Dept. of IT Convergence & Applications Engineering, Pukyong National University)
  • 투고 : 2021.12.20
  • 심사 : 2021.12.30
  • 발행 : 2021.12.31

초록

용접 비드의 불량유무를 판단하기 위하여 다양한 방법들이 제안되어왔으며, 최근에는 센서 자료 검사와 영상 자료 검사가 꾸준히 발표되고 있다. 센서 자료 검사는 정확도가 높고, 2차원 기반의 영상 자료 검사는 용접된 비드의 위치를 파악할 수 있다는 장점이 있다. 하지만 센서 자료만으로 분석 할 경우 정확한 위치에 용접이 되었는지 파악하기가 어렵다. 반면 영상 자료 방법은 잡음과 측정오차가 발생하여 정확도가 높지 않다. 본 논문에서는 평균 전압, 평균 전류, 혼합가스인 센서 자료 검사와 영상 검사 방법을 융합함으로써 각 검사 방법들의 단점을 보완하고 장점을 높여서 정확도를 향상시키고 검사 속도를 높일 수 있는 방법과 이를 소프트웨어로 구현하였다. 그리고 그래픽 사용자 인터페이스(graphical user interface; GUI)를 이용하여 분석을 수행하고 검사에 사용된 자료와 검사 결과를 확인할 수 있도록 하여 사용자가 편리하고 직관적으로 분석을 수행하고 결과를 파악할 수 있도록 하고자 한다. 이 때 각 센서 자료의 특성을 이용하여 센서 검사가 수행되고, 모폴로지 축지적 활성화 윤곽선을 적용하여 영상 자료가 검사된다. 실험 결과를 통하여 98%의 정확도를 보였으며, 네 개의 용접 영상과 센서 자료 검사를 모두 수행할 경우 검사 시간은 약 1.9초로서 용접공정에서 실시간 검사기로 사용 가능한 소프트웨어의 성능을 보였다.

Various methods have been proposed to determine the defect detection of welding bead, and recently sensor data and image data inspection have been steadily announced. There are advantages that sensor data inspection is highly accurate, and two-dimensional-based image data inspection is able to determine the position of the welding bead. However, when analyzing only with sensor data, it is difficult to determine whether the welding has been performed at the correct position. On the other hand, the image data inspection does not have high accuracy due to noise and measurement errors. In this paper, we propose a method that can complement the shortcomings of each inspection method and increase its advantages to improve accuracy and speed up inspection by fusing sensor data inspection which are average current, average volt, and mixed gas data, and image data inspection methods and is implemented as software. In addition, it is intended to allow users to conveniently and intuitively analyze and grasp the results by performing analysis using a graphical user interface(GUI) and checking the data and inspection results used for the inspection. Sensor inspection is performed using the characteristics of each sensor data, and image data is inspected by applying a morphology geodesic active contour algorithm. The experimental results showed 98% accuracy, and when performing the inspection on the four image data, and sensor data the inspection time was about 1.9 seconds, indicating the performance of software that can be used as a real-time inspector in the welding process.

키워드

과제정보

본 논문은 과학기술정보통신부 지역SW서비스 사업화 지원과제, 한국연구재단 기초연구사업, 산업통상자원부와 한국산업기술진흥원 지역혁신클러스터 육성사업(R&D P0004797)의 지원으로 수행되었습니다.

참고문헌

  1. M. S. Kim, S. M. Shin, D. H. Kim, and S. h. Rhee. "A study on the algorithm for determining back bead generation in GMA welding using deep learning," Journal of Welding and Joining, Vol. 36, No. 2, pp. 74-81, 2018. https://doi.org/10.5781/jwj.2018.36.2.11
  2. J. H. Kang, and N. K. Ku. "Verification of Resistance Welding Quality Based on Deep Learning," Journal of the Society of Naval Architects of Korea, Vol. 56, No. 6, pp. 473-479, 2019. https://doi.org/10.3744/SNAK.2019.56.6.473
  3. J. H. Kim, M.H. Ko, N. K. Ku. "A Study on Resistance Spot Welding Failure Detection Using Deep Learning Technology," Journal of Computational Design and Eng, Vol. 24, No. 2, pp. 161-168, 2019.
  4. L. Alvarez, L. Baumela, P. Henriquez, and P. M. Neila, "Morphological Snakes," Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Francisco, CA, pp. 2197-2202, 2010.
  5. Z. Wang, and Y. Liu, "Active contour model by combining edge and region information discrete dynamic systems," Advances in Mechanical Engineering Vol. 9, No. 3, pp. 1-10, 2017.
  6. J. Ma, D. Wang, X. P. Wang, and X. Yang, "A Fast Algorithm for Geodesic Active Contours with Applications to Medical Image Segmentation," Computer Vision and Pattern Recognition pp. 1-10, 2020.
  7. L. Alvarez, L. Baumela, P. M. Neila, and P. Henriquez, "A Real Time Morphological Snakes Algorithm," Image Processing on Line, Vol. 2, pp. 1-7, 2014. https://doi.org/10.5201/ipol.2012.abmh-rtmsa
  8. X.X.,Yin, Y. Jian, Y. Zhang, J. Wu, H. Lu, and M. Y. Su. "Automatic breast tissue segmentation in MRIs with morphology snake and deep denoiser training via extended Stein's unbiased risk estimator." Health Information Science and Systems, Vol.9, No.1, pp.1-21, 2021.
  9. O. Haffner, E. Kucera, and S. Kozak, "Weld Segmentation for Diagnostic and Evaluation Method," Cybernetics & Informatics, Levoca, Slovakia, pp. 1-6, 2016.
  10. J. Mlyahilu,Y. B Kim, J. E Lee and J. N. Kim. An Algorithm of Welding Bead Detection and Evaluation Using and Multiple Filters Geodesic Active Contour. The Journal of Korea Institute of Convergence Signal Processing, Vol.22, No.3, pp.141-148, 2021. https://doi.org/10.23087/JKICSP.2021.22.3.009
  11. J. W. Baek, and S. G. Kong, "Faster R-CNN Classifier Structure Design for Detecting Weld Surface Defects," In Proceedings of KIIT Conference, pp. 149-152, 2018.
  12. O. Haffner, E. Kucera, P. Drahos, and J. Ciganek, "Using Entropy for Welds Segmentation and Evaluation. Entropy," Vol.21, No.12, pp.1168, 2019. https://doi.org/10.3390/e21121168
  13. T. W. Kim, C. M. Han, and H. W. Choi, " Welding Quality Inspection using Deep Learning," The Korean Society of Manufacturing Process Engineers, pp. 18-18, 2020.