Automotive Experiences

Articles

Visual Inspection Improvement of Engine Components Using Deep Learning with Pre-processed Dataset Augmentation: Case Study

Steven Salim , Sandy Wiratama , Alfan Sarifudin , Eka Prita Yuliatin

Published
December 5, 2025

Vol. 8 No. 3 (2025)
Pages 512-529

DOI

Open Access
Share

Abstract

Lighting instability, sharp shadows, and visual disturbances caused by mechanical vibrations are significant challenges in the application of computer vision-based visual inspection systems in automotive industrial environments. This study aims to enhance the accuracy and robustness of the YOLOv8 object detection model for detecting machine component completeness by applying an adaptive pre-processing strategy. The techniques employed include grayscale conversion, brightness adjustment, and blurring to simulate common visual conditions encountered in real-world production processes. The model was trained using 1,281 instances from 52 component classes and evaluated based on the metrics of precision, recall, mAP@50, and mAP@50”“95. The results show an average precision of 0.971, a recall of 0.990, and mAP@50 of 0.991, with spatial variation reflected in the standard deviation of mAP@50”“95 of 0.149. The pre-processing technique improves the detection precision of shape-based components by up to 19% and colour-based components by up to 31%. Testing on ten appearance variations showed 100% detection accuracy with no misclassification, indicating the model’s generalizability to data in the training distribution. These findings confirm that visual modification of training data significantly improves the reliability and efficiency of the YOLOv8-based automated inspection system. Further implications include reduced human intervention, accelerated production flow, and optimization of operational energy consumption through faster and more accurate detection. Therefore, this system contributes to energy-efficient and sustainable innovative manufacturing practices.

Keywords

Object detection; Automatic inspection; Automotive factory intelligence; Engine components; Smart manufacturing; Sustainable production

References

  1. S. Grzesiak and A. Sulich, “Car Engines Comparative Analysis: Sustainable Approach,†Energies, vol. 15, no. 14, p. 5170, Jul. 2022, doi: 10.3390/en15145170.
  2. M. Mazzetto, M. Teixeira, É. O. Rodrigues, and D. Casanova, “Deep Learning Models for Visual Inspection on Automotive Assembling Line,†Int. J. Adv. Eng. Res. Sci., vol. 7, no. 3, pp. 473–494, 2020, doi: 10.22161/ijaers.74.56.
  3. U. Sirisha, S. P. Praveen, P. N. Srinivasu, P. Barsocchi, and A. K. Bhoi, “Statistical Analysis of Design Aspects of Various YOLO-Based Deep Learning Models for Object Detection,†Int. J. Comput. Intell. Syst., vol. 16, no. 1, p. 126, Aug. 2023, doi: 10.1007/s44196-023-00302-w.
  4. M. Hussain, “YOLO-v1 to YOLO-v8, the Rise of YOLO and Its Complementary Nature toward Digital Manufacturing and Industrial Defect Detection,†Machines, vol. 11, no. 7, p. 677, Jun. 2023, doi: 10.3390/machines11070677.
  5. M. Safaldin, N. Zaghden, and M. Mejdoub, “An Improved YOLOv8 to Detect Moving Objects,†IEEE Access, vol. 12, pp. 59782–59806, 2024, doi: 10.1109/ACCESS.2024.3393835.
  6. X. Liu, C. Liu, X. Zhou, and G. Fan, “Enhancing low-light object detection with En-YOLO: leveraging dual attention and implicit feature learning,†Multimed. Syst., vol. 31, no. 3, p. 249, Jun. 2025, doi: 10.1007/s00530-025-01820-7.
  7. Z. Sun et al., “An Anchor-Free Detection Method for Ship Targets in High-Resolution SAR Images,†IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens., vol. 14, pp. 7799–7816, 2021, doi: 10.1109/JSTARS.2021.3099483.
  8. M. Yeung, E. Sala, C.-B. Schönlieb, and L. Rundo, “Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation,†Comput. Med. Imaging Graph., vol. 95, p. 102026, Jan. 2022, doi: 10.1016/j.compmedimag.2021.102026.
  9. X. Xu et al., “Crack Detection and Comparison Study Based on Faster R-CNN and Mask R-CNN,†Sensors, vol. 22, no. 3, p. 1215, Feb. 2022, doi: 10.3390/s22031215.
  10. D. Devi, S. Sophia, and S. R. Boselin Prabhu, “Deep learning-based cognitive state prediction analysis using brain wave signal,†in Cognitive Computing for Human-Robot Interaction, Elsevier, 2021, pp. 69–84. doi: 10.1016/B978-0-323-85769-7.00017-3.
  11. D. Wan et al., “YOLO-MIF: Improved YOLOv8 with Multi-Information fusion for object detection in Gray-Scale images,†Adv. Eng. Informatics, vol. 62, p. 102709, Oct. 2024, doi: 10.1016/j.aei.2024.102709.
  12. I. Zeger, S. Grgic, J. Vukovic, and G. Sisul, “Grayscale Image Colorization Methods: Overview and Evaluation,†IEEE Access, vol. 9, pp. 113326–113346, 2021, doi: 10.1109/ACCESS.2021.3104515.
  13. J. Wang and S. Lee, “Data Augmentation Methods Applying Grayscale Images for Convolutional Neural Networks in Machine Vision,†Appl. Sci., vol. 11, no. 15, p. 6721, Jul. 2021, doi: 10.3390/app11156721.
  14. Y. Ge, Q. Zhang, Y. Sun, Y. Shen, and X. Wang, “Grayscale medical image segmentation method based on 2D&3D object detection with deep learning,†BMC Med. Imaging, vol. 22, no. 1, p. 33, Dec. 2022, doi: 10.1186/s12880-022-00760-2.
  15. S. Wang, “Evaluation of impact of image augmentation techniques on two tasks: Window detection and window states detection,†Results Eng., vol. 24, p. 103571, Dec. 2024, doi: 10.1016/j.rineng.2024.103571.
  16. X. Liu, H. Liu, and D. Zhou, “Multi-scale defect detection on product surface in grayscale images using YOLOv8-Ms algorithm,†Nondestruct. Test. Eval., pp. 1–25, Mar. 2025, doi: 10.1080/10589759.2025.2478259.
  17. C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,†J. Big Data, vol. 6, no. 1, p. 60, Dec. 2019, doi: 10.1186/s40537-019-0197-0.
  18. G.-J. Son, H.-C. Jung, and Y.-D. Kim, “Temporal-Quality Ensemble Technique for Handling Image Blur in Packaging Defect Inspection,†Sensors, vol. 24, no. 14, p. 4438, Jul. 2024, doi: 10.3390/s24144438.
  19. M. D. Cirillo, D. Abramian, and A. Eklund, “What is The Best Data Augmentation For 3D Brain Tumor Segmentation?,†in 2021 IEEE International Conference on Image Processing (ICIP), IEEE, Sep. 2021, pp. 36–40. doi: 10.1109/ICIP42928.2021.9506328.
  20. K. Chao, W. Song, S. Shao, D. Liu, X. Liu, and X. Zhao, “CUI-Net: a correcting uneven illumination net for low-light image enhancement,†Sci. Rep., vol. 13, no. 1, p. 12894, Aug. 2023, doi: 10.1038/s41598-023-39524-5.
  21. C.-T. Chien, R.-Y. Ju, K.-Y. Chou, E. Xieerke, and J.-S. Chiang, “YOLOv8-AM: YOLOv8 Based on Effective Attention Mechanisms for Pediatric Wrist Fracture Detection,†IEEE Access, vol. 13, pp. 52461–52477, 2025, doi: 10.1109/ACCESS.2025.3549839.
  22. Z.-Q. Zhao, P. Zheng, S.-T. Xu, and X. Wu, “Object Detection With Deep Learning: A Review,†IEEE Trans. Neural Networks Learn. Syst., vol. 30, no. 11, pp. 3212–3232, Nov. 2019, doi: 10.1109/TNNLS.2018.2876865.
  23. G. M. Foody, “Challenges in the real world use of classification accuracy metrics: From recall and precision to the Matthews correlation coefficient,†PLoS One, vol. 18, no. 10, p. e0291908, Oct. 2023, doi: 10.1371/journal.pone.0291908.
  24. T.-Y. Lin et al., “Microsoft COCO: Common Objects in Context,†2014, pp. 740–755. doi: 10.1007/978-3-319-10602-1_48.
  25. L. He, Y. Zhou, L. Liu, W. Cao, and J. Ma, “Research on object detection and recognition in remote sensing images based on YOLOv11,†Sci. Rep., vol. 15, no. 1, p. 14032, Apr. 2025, doi: 10.1038/s41598-025-96314-x.
  26. Z. Yuan et al., “Research on Calf Behavior Recognition Based on Improved Lightweight YOLOv8 in Farming Scenarios,†Animals, vol. 15, no. 6, p. 898, Mar. 2025, doi: 10.3390/ani15060898.
  27. N. Hütten, M. Alves Gomes, F. Hölken, K. Andricevic, R. Meyes, and T. Meisen, “Deep Learning for Automated Visual Inspection in Manufacturing and Maintenance: A Survey of Open- Access Papers,†Appl. Syst. Innov., vol. 7, no. 1, p. 11, Jan. 2024, doi: 10.3390/asi7010011.
Article Metrics
20
Views
32
Downloads

SCITE

DIMENSIONS

Publication Information
Published December 5, 2025
Section Articles
Copyright Copyright (c) 2025 Steven Salim, Sandy Wiratama, Alfan Sarifudin, Eka Prita Yuliatin
Keywords
  • Object detection
  • Automatic inspection
  • Automotive factory intelligence
  • Engine components
  • Smart manufacturing
  • Sustainable production