Article Details

Title · Authors · Abstract · Keywords · DOI · References

A Lightweight Algorithm for Stacked Timber Anomaly Detection on Construction Sites Based on Improved YOLOv8n

WANG Haoyu

(College of Computer Science and Technology, Taiyuan Normal University, Jinzhong 030619, China)

Section：Artificial Intelligence · Vol. 2025, Issue 7 · Pages：58-63,70 · 2096-4706（2025）07-0058-07 · CLC number：TP183;TP391 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.07.011

Abstract

When the timber materials are stacked on the construction site, the outdoor environment is prone to abnormal problems such as moisture deformation and dry cracking on the surface of the timber. Aiming at the problems of poor accuracy and high computational complexity of the existing detection algorithms on the surface of timber materials, a lightweight small target detection algorithm (YOLO-ESN) based on YOLOv8n is proposed. The algorithm introduces the Spatial and Channel Reconstruction Convolution (SCConv) module and the Normalized Wasserstein Distance (NWD) loss function for small target detection. At the same time, it embeds the Efficient Multi-Scale Attention (EMA) module based on Cross-Spatial Learning into the backbone network to reduce the impact of occlusion and background interference. The improved algorithm is experimentally verified on the timber defect dataset. Compared with the original algorithm, its mAP@0.5 is increased by 3.6%, and the parameter quantity is reduced by 23.3%, which realizes the real-time and accurate detection of the abnormal situation of stacked timber materials.

Keywords

improved YOLOv8n algorithm; construction site timber anomaly detection; lightweight; small target detection

References

  1. [1] Feng Lijun Analysis of the Causes of Construction Safety Accidents and Research on Early Warning Management [D]. Tianjin: Tianjin University of Finance and Economics, 2008
  2. [2] Pei Wenbo, Wang Shuilian Analysis of Common Safety Hazards in Construction Sites and Suggestions for Countermeasures [J]. Building Safety, 2020, 35 (10): 52-55
  3. [3] QAYYUM R，KAMAL K，ZAFAR T，et al. Wood Defects Classification Using GLCM Based Features and PSO Trained Neural Network [C]//2016 22nd International Conference on Automation and Computing (ICAC).Colchester：IEEE，2016：273-277.
  4. [4] AGRAWAL P，GIRSHICK R，MALIK J. Analyzing the Performance of Multilayer Neural Networks for Object Recognition [C]//Computer Vision-ECCV 2014.Zurich：Springer，2014：329-344.
  5. [5] GIRSHICK R. Fast R-CNN [C]//2015 IEEE International Conference on Computer Vision (ICCV).Santiago：IEEE，2015：1440-1448.
  6. [6] REN S Q，HE K M，GIRSHICK R，et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks [C]//IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，39（6）：1137-1149.
  7. [7] REDMON J，DIVVALA S，GIRSHICK R，et al. You Only Look Once: Unified， Real-Time Object Detection [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas：IEEE，2016：779-788.
  8. [8] LIU W，ANGUELOV D，ERHAN D，et al. SSD: Single Shot MultiBox Detector [C]//Computer Vision-ECCV 2016.Amsterdam：Springer，2016：21-37.
  9. [9] SERMANET P，EIGEN D，ZHANG X，et al. OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks [J/OL]. arXiv:1312.6229 [cs.CV].[2024-10-16].https://doi.org/10.48550/arXiv.1312.6229.
  10. [10] Chen Junyu Exploring the Application of Smart Construction Sites in Safety and Quality Management of Construction Projects [J]. Product Reliability Report, 2023 (8): 45-46
  11. [11] Han Qiang Research on Improved YOLOv8 Algorithm for Small Object Detection [D]. Changchun: Jilin University, 2023
  12. [12] OUYANG D L，HE S，ZHANG G Z，et al. Efficient Multi-Scale Attention Module with Cross-Spatial Learning [C]//CASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).Rhodes Island：IEEE，2023：1-5.
  13. [13] WANG J W，XU C，YANG W，et al. A Normalized Gaussian Wasserstein Distance for Tiny Object Detection [J/OL].arXiv:2110.13389 [cs.CV].[2024-10-23].https://doi.org/10.48550/arXiv.2110.13389.
  14. [14] LI J F，WEN Y，HE L H，et al. SCConv: Spatial and Channel Reconstruction Convolution for Feature Redundancy [C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Vancouver：IEEE，2023：6153-6162.
  15. [15] CHEN Y P，DAI X Y，LIU M C，et al. Dynamic Convolution: Attention Over Convolution Kernels [C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Seattle：IEEE，2020：11027-11036.
  16. [16] HU J，SHEN L，SUN G. Squeeze-and-Excitation Networks [C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City：IEEE，2018：7132-7141.
  17. [17] WOO S，PARK J，LEE J Y，et al. CBAM: Convolutional Block Attention Module [J/OL].arXiv:1807.06521 [cs.CV].[2024-09-28].https://doi.org/10.48550/arXiv.1807.06521.
  18. [18] IANDOLA F N，HAN S，MOSKEWICZ M W，et al. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and＜0.5 MB model size [J/OL].arXiv:1602.07360 [cs.CV].[2024-09-12].https://doi.org/10.48550/arXiv.1602.07360.
  19. [19] HOWARD A G，ZHU M L，CHEN B，et al. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications [J/OL].arXiv:1704.04861 [cs.CV].[2024-09-10].https://doi.org/10.48550/arXiv.1704.04861.

About the authors

Wang Haoyu (1996-), male, Han ethnicity, from Changzhi, Shanxi Province, is currently pursuing a master's degree in image processing.