Publisher

IAESONLINEIAESONLINE

Journal Indonesian Journal of Electrical Engineering and Informatics (IJEEI)Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Abstract

This study examines how deep learning models can improve corrosion detection, comparing YOLOv7 with its more advanced version, YOLOv8. Both models were trained on a diverse set of images showing different types and levels of corrosion on metal surfaces. Their performance was assessed using standard industry metrics, including accuracy, F1-score, recall, and mean average precision (mAP). The results clearly show that YOLOv8 outperforms YOLOv7 in all areas. It achieves higher recall, precision, and F1-score, demonstrating its improved ability to detect and classify corroded areas. Notably, YOLOv8 is better at identifying small or early-stage corrosion, which is crucial for timely maintenance. Additionally, it processes images faster than YOLOv7, making it more suitable for real-time applications. This study also suggests integrating YOLOv8 with robotic arms equipped for laser cleaning, allowing for automated and precise corrosion removal. This system could improve maintenance efficiency, reduce costs, and enhance the safety and reliability of infrastructure.

Conclusion

In conclusion, this study demonstrated the superiority of YOLOv8 over YOLOv7 in accurately and efficiently detecting corrosion in real-world images.YOLOv8 outperformed YOLOv7 in key performance metrics such as recall, precision, and F1-score, and also exhibited superior performance in detecting small corrosion areas and faster inference times.The implementation of corrosion detection systems can revolutionize inspection practices in sectors relying on metallic infrastructure, enhancing safety, saving costs, and increasing efficiency.

Future Research

Future research should explore methods to improve the detection of small and subtle corrosion defects, potentially through the integration of additional sensor data or advanced image preprocessing techniques. Furthermore, investigating the use of contextual information, such as environmental factors and material properties, could enhance the models understanding of corrosion patterns and improve its predictive capabilities. Finally, a promising avenue for future work involves developing adaptive learning algorithms that can continuously refine the detection model based on real-world feedback, enabling it to effectively address the evolving challenges of corrosion monitoring in diverse industrial settings. These advancements will contribute to more robust, reliable, and cost-effective corrosion management systems, ultimately extending the lifespan and safety of critical infrastructure. The development of such systems requires a multidisciplinary approach, combining expertise in deep learning, materials science, and engineering to address the complex challenges of corrosion detection and prevention.

References

1. Deep Learning for Entity Matching | Proceedings of the 2018 International Conference on Management of... doi.org/10.1145/3183713.3196926Deep Learning for Entity Matching Proceedings of the 2018 International Conference on Management of doi 10 1145 3183713 3196926
2. Efficient Corrosion Detection on Metal Surface Using Deep Learning Technique | Ky Phuc | Indonesian Journal... doi.org/10.52549/ijeei.v13i3.6420Efficient Corrosion Detection on Metal Surface Using Deep Learning Technique Ky Phuc Indonesian Journal doi 10 52549 ijeei v13i3 6420
3. [1506.02640] You Only Look Once: Unified, Real-Time Object Detection. look once unified real time object... doi.org/10.48550/arXiv.1506.026401506 02640 You Only Look Once Unified Real Time Object Detection look once unified real time object doi 10 48550 arXiv 1506 02640

Metadata

File size	1.31 MB
Pages	11
DMCA	Report