International conference on construction engineering and project management (국제학술발표논문집)

  • 2024.07a
  • /
  • Pages.311-318
  • /
  • 2024
  • /
  • 2508-9048(eISSN)
Korea Institute of Construction Engineering and Management (한국건설관리학회)
권호 표지
DOI QR코드

DOI QR Code

HYBRID DATA SET GENERATION METHOD FOR COMPUTER VISION-BASED DEFECT DETECTION IN BUILDING CONSTRUCTION

  • Seung-mo Choi (Department of Architectural Engineering, Ajou University) ;
  • Heesung Cha (Department of Architectural Engineering, Ajou University) ;
  • Bo-sik, Son (Department of Architectural Engineering, Namseoul University)
  • Published : 2024.07.29
Download PDF
⟨ Previous Next ⟩

Abstract

Quality control in construction projects necessitates the detection of defects during construction. Currently, this task is performed manually by site supervisors. This manual process is inefficient, labor-intensive, and prone to human error, potentially leading to decreased productivity. To address this issue, research has been conducted to automate defect detection using computer vision-based object detection technologies. However, these studies often suffer from a lack of data for training deep learning models, resulting in inadequate accuracy. This study proposes a method to improve the accuracy of deep learning models through the use of virtual image data. The target building is created as a 3D model and finished with materials similar to actual components. Subsequently, a virtual defect texture is produced by layering three types of images: defect information, area information, and material information images, to fabricate materials with defects. Images are generated by rendering the 3D model and the defect, and annotations are created for segmentation. This approach creates a hybrid dataset by combining virtual data with actual site image data, which is then used to train the deep learning model. This research was conducted on the tile process of finishing construction projects, focusing on cracks and falls as the target defects. The training results of the deep learning model show that the F1-Score increased by 12.08% for falls and cracks when using the hybrid dataset compared to the real image dataset alone, validating the hybrid data approach. This study contributes not only to unmanned and automated smart construction management but also to enhancing safety on construction sites. To establish an integrated smart quality management system, it is necessary to detect various defects simultaneously with high accuracy. Utilizing this method for automatic defect detection in other types of construction can potentially expand the possibilities for implementing an integrated smart quality management system.

Keywords

Acknowledgement

This work was supported by the Ajou University Research Fund.

References

  1. D. Hui, H. Hao, L. Dehuan, D. Yichuan, S. Cheng, Automatic Indoor Construction Process Monitoring for Tiles Based on BIM and Computer Vision, J. Constr. Eng. Manage. 146 (2020) 04019095.
  2. P. Martinez, M. Al-Hussein, R. Ahmad, A scientometric analysis and critical review of computer vision applications for construction, Autom. Constr. 107 (2019) 102947.
  3. H. Perez, J.H. Tah, Deep learning smartphone application for real-time detection of defects in buildings, Structural Control and Health Monitoring. 28 (2021) e2751.
  4. D. Thikra, Z. Zhenhua, Z. Tarek, Computer Vision-Based Model for Moisture Marks Detection and Recognition in Subway Networks, J. Comput. Civ. Eng. 32 (2018) 04017079.
  5. H. Luo, L. Lin, K. Chen, M.F. Antwi-Afari, L. Chen, Digital technology for quality management in construction: A review and future research directions, Developments in the Built Environment. 12 (2022) 100087.
  6. Z. Gong, P. Zhong, W. Hu, Diversity in machine learning, Ieee Access. 7 (2019) 64323-64350.
  7. F.E. Nowruzi, P. Kapoor, D. Kolhatkar, F.A. Hassanat, R. Laganiere, J. Rebut, How much real data do we actually need: Analyzing object detection performance using synthetic and real data, arXiv preprint arXiv:1907.07061. (2019).
  8. D. Kim, The priority analysis of defect type through tenant preliminary research in apartment house, Master's Thesis, Korea University. (2016).
  9. H. Ying, R. Sacks, A. Degani, Synthetic image data generation using BIM and computer graphics for building scene understanding, Autom. Constr. 154 (2023) 105016.
  10. K. Park, S. Ergan, C. Feng, Quality assessment of residential layout designs generated by relational Generative Adversarial Networks (GANs), Autom. Constr. 158 (2024) 105243.
  11. L. Eversberg, J. Lambrecht, Generating images with physics-based rendering for an industrial object detection task: Realism versus domain randomization, Sensors. 21 (2021) 7901.
  12. J. Kim, D. Kim, S. Lee, S. Chi, Hybrid DNN training using both synthetic and real construction images to overcome training data shortage, Autom. Constr. 149 (2023) 104771.
  13. T. Hodan, V. Vineet, R. Gal, E. Shalev, J. Hanzelka, T. Connell, P. Urbina, S.N. Sinha, B. Guenter, Photorealistic image synthesis for object instance detection, (2019) 66-70.
  14. Y. Hong, S. Park, H. Kim, H. Kim, Synthetic data generation using building information models, Autom. Constr. 130 (2021) 103871.
  15. H. Lee, J. Jeon, D. Lee, C. Park, J. Kim, D. Lee, Game engine-driven synthetic data generation for computer vision-based safety monitoring of construction workers, Autom. Constr. 155 (2023) 105060.
  16. C. Siu, M. Wang, J.C.P. Cheng, A framework for synthetic image generation and augmentation for improving automatic sewer pipe defect detection, Autom. Constr. 137 (2022) 104213.
  17. T. Tschickardt, F. Kaufmann, C. Glock, Lean and BIM based flight planning for automated data acquisition of bridge structures with LiDAR UAV during construction phase, 3 (2022) 0.
  18. P. Mittal, R. Singh, A. Sharma, Deep learning-based object detection in low-altitude UAV datasets: A survey, Image Vision Comput. 104 (2020) 104046.
  19. D. Wallace, Y.H. He, J.C. Vaz, L. Georgescu, P.Y. Oh, Multimodal teleoperation of heterogeneous robots within a construction environment, (2020) 2698-2705.
  20. Y. Tan, W. Yi, P. Chen, Y. Zou, An adaptive crack inspection method for building surface based on BIM, UAV and edge computing, Autom. Constr. 157 (2024) 105161.