Design of YOLOv5 Medium as Unmanned Rail Inspection in Braking Control System Based on Computer Vision

Adiratna Ciptaningrum, Andhika Putra Widyadharma, Imam Junaedi, Rahayu Mekar Bisono, R. Akbar Nur Apriyanto, R. Gaguk Pratama Yudha, Mohammad Erik Echsony, Larissa Kartika Putri, Edo Zulmi Faikhsan

Abstract

This research explores the application of computer vision using YOLOv5 Medium for automatic object detection in unmanned rail inspection systems. The proposed technique utilizes image processing to analysing pixel values and detect optical motion vehicles. This detection triggers control system responses, such as activating the inspection train's motor upon vehicle identification. The study demonstrates the effectiveness of YOLOv5 Medium in achieving high accuracy rates. Evaluations at various distances yielded promising results: 97.98% at 3 meters, 100% at 5 meters, 99.49% at 7 meters, and a perfect 100% at 9 meters. These findings suggest optimal system performance at a distance of 9 meters. Overall detection performance across all test distances remained consistently high, with sequential rates of 0.96, 0.97, 0.95, and 0.96. This research emphasizes the crucial role of several factors in maintaining system accuracy and performance. These include the efficacy of the colour segmentation algorithm, ambient lighting conditions, and camera resolution. Furthermore, the importance of extensive testing with a diverse dataset is highlighted to ensure the system's robustness and adaptability to various real-world scenarios.

Keywords

unmanned rail inspection, object detection, YOLOv5 Medium

Full Text:

PDF

References

S. Sumit, D. Rambli, and A. Mirjalili, “Vision-Based Human Detection Techniques: A Descriptive Review,” IEEE Access, vol. 9, January 2021.

Viola, P., & Jones, M. J. (2001). Rapid object detection using a boosted cascade of simple features. In Proceedings of the 2001 IEEE computer society conference on computer vision and pattern recognition (CVPR) (Vol. I, pp. I-511-I-518). IEEE.

Dalal, N., & Triggs, B. (2005). Histogram of oriented gradients for human detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR) (Vol. 1, pp. 886-893).

Felzenszwalb, P. F., Girshick, R. B., David, D., & Malik, J. (2010). Object detection with deformable part models. IEEE transactions on pattern analysis and machine intelligence, 32(8), 1407-1417.

Zhao, Q., Pang, Y., Fu, H., Li, Z., & Xu, D. (2023). Recent Advances in Deep Learning-Based Object Detection. IEEE Access, 11, 28238-28264.

D. Zhu, L. Dai and P. Du, "CCE-YOLOv5s: An Improved YOLOv5 Model for UAV Small Target Detection," 2023 IEEE 5th International Conference on Civil Aviation Safety and Information Technology (ICCASIT), Dali, China, 2023, pp. 824-829

S. Biswas, A. Nandy, A. K. Naskar and R. Saw, "Real time Gesture Recognition using Improved YOLOv5 Model," 2024 11th International Conference on Signal Processing and Integrated Networks (SPIN), Noida, India, 2024, pp. 328-333

B. Zheng, P. Li and H. Lu, "Research on Moving Target Detection Technology Based on Yolov5 Model," 2023 2nd International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP), Hangzhou, China, 2023, pp. 305-309

DOI

https://doi.org/10.21107/ijseit.v8i2.26199

Metrics

Refbacks

  • There are currently no refbacks.


Copyright (c) 2024 Adiratna Ciptaningrum, Andhika Putra Widyadharma, Imam Junaedi, Rahayu Mekar Bisono, R. Akbar Nur Apriyanto, R. Gaguk Pratama Yudha, Mohammad Erik Echsony, Larissa Kartika Putri, Edo Zulmi Faikhsan

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.