Inspection System for Vehicle Headlight Defects Based on Convolutional Neural Network
Source abstract
Abstract
This paper proposes a method to detect the defects in the region of interest (ROI) based on a convolutional neural network (CNN) after alignment (position and rotation calibration) of a manufacturer’s headlights to determine whether the vehicle headlights are defective. The results were compared with an existing method for distinguishing defects among the previously proposed methods. One hundred original headlight images were acquired for each of the two vehicle types for the purpose of this experiment, and 20,000 high quality images and 20,000 defective images were obtained by applying the position and rotation transformation to the original images. It was found that the method proposed in this paper demonstrated a performance improvement of more than 0.1569 (15.69% on average) as compared to the existing method.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Advanced Measurement and Detection Methods · Optical measurement and interference techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
ICT-Convergence Research Center, Kumoh National Institute of Technology, Gumi 39177, Korea
Computer and Software Engineering, Kumoh National Institute of Technology, Gumi 39177, Korea
Department of IT Covergence Engineering, Kumoh National Institute of Technology, Gumi 39177, Korea
Funding
IITP · NRF
References (19)
Show source-provided references
1. A Fast Way for Alignment Marker Detection and Position Calibration10.3745/KTSDE.2016.5.1.35 ↗
2. Detecting Pedestrians using Patterns of Motion and Appearance10.1007/s11263-005-6644-8 ↗
3. Object Detection with Discriminatively Trained Part-based Models10.1109/TPAMI.2009.167 ↗
4. Learning Patterns of Activity using Real-time Tracking10.1109/34.868677 ↗
5. Imagenet Classification with Deep Convolutional Neural Networks
6. Lu, C.-C., and Juang, J.-G. (2020). Robotic-Based Touch Panel Test System Using Pattern Recognition Methods. Appl. Sci., 10.10.3390/app10238339 ↗
7. Implementation of Automatic Detection System for CCFL’s Defects based on Combined Lighting
8. Matuszek, J., Seneta, T., and Moczała, A. (2020). Assessment of the Design for Manufacturability Using Fuzzy Logic. Appl. Sci., 10.10.20944/preprints202003.0396.v1 ↗
9. Kinematics and control of a visual alignment system for flat panel displays10.5302/J.ICROS.2008.14.4.369 ↗
10. Park, C.S., and Kwon, S.J. (2006, January 18–21). An efficient vision algorithm for fast and fine mask-panel alignment. Proceedings of the 2006 SICEICASE International Joint Conference (SICE-ICCAS 2006), Busan, Korea.10.1109/SICE.2006.315757 ↗
11. Jung, U.-K., Moon, C.B., Lee, H.-Y., Kim, B.M., and Yang, H.S. (2009, January 27). Preprocessing Algorithm for Detecting CCFL Defects. Proceedings of the 2009 Conference on Korean Institute of Information Scientists and Engineers, Seoul, Korea.
12. Inspection of Vehicle Headlight Defects
13. (2021, January 04). Condensation. Available online: http://opencv.jp/sample/estimators.html.
14. (2021, January 04). Particle Filter Tracking in Python. Available online: https://www.slideshare.net/kohta/particle-filter-tracking-in-python.
15. Pulli, K., Baksheev, A., Kornyakov, K., and Eruhimov, V. (2012). Realtime Computer Vision with OpenCV, ACM Queue.10.1145/2181796.2206309 ↗
16. Hwang, S. (2007). Image Processing Programming by Visual C++, Hanbit Publishing Network.
17. Shane, T. (2016). Applied Computer Science, Springer. [2nd ed.].
18. Kim, B.M., Lee, J.Y., and Moon, C.B. (2020). Deep Learning in Practice, Hongreung Publishing Company.
19. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., and Rabinovich, A. (2014, January 23–28). Going Deeper with Convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.10.1109/CVPR.2015.7298594 ↗