The Efficiency of YOLOv5 Models in the Detection of Similar Construction Details
Source abstract
Abstract
Computer vision solutions have become widely used in various industries and as part of daily solutions. One task of computer vision is object detection. With the development of object detection algorithms and the growing number of various kinds of image data, different problems arise in relation to the building of models suitable for various solutions. This paper investigates the influence of parameters used in the training process involved in detecting similar kinds of objects, i.e., the hyperparameters of the algorithm and the training parameters. This experimental investigation focuses on the widely used YOLOv5 algorithm and analyses the performance of different models of YOLOv5 (n, s, m, l, x). In the research, the newly collected construction details (22 categories) dataset is used. Experiments are performed using pre-trained models of the YOLOv5. A total of 185 YOLOv5 models are trained and evaluated. All models are tested on 3300 images photographed on three different backgrounds: mixed, neutral, and white. Additionally, the best-obtained models are evaluated using 150 new images, each of which has several dozen construction details and is photographed against different backgrounds. The deep analysis of different YOLOv5 models and the hyperparameters shows the influence of various parameters when analysing the object detection of similar objects. The best model was obtained when the YOLOv5l was used and the parameters are as follows: coloured images, image size—320; batch size—32; epoch number—300; layers freeze option—10; data augmentation—on; learning rate—0.001; momentum—0.95; and weight decay—0.0007. These results may be useful for various tasks in which small and similar objects are analysed.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: 3D Surveying and Cultural Heritage · Image Processing and 3D Reconstruction
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Information Technology, Vilnius Gediminas Technical University, Saulėtekio al. 11, LT-10223 Vilnius, Lithuania
Department of Electronic Systems, Vilnius Gediminas Technical University, Saulėtekio al. 11, LT-10223 Vilnius, Lithuania
Department of Information Systems, Vilnius Gediminas Technical University, Saulėtekio al. 11, LT-10223 Vilnius, Lithuania
References (44)
Show source-provided references
1. Jucevičius, J., Treigys, P., Bernatavičienė, J., Briedienė, R., Naruševičiūtė, I., and Trakymas, M. (2021, January 22–24). Investigation of MRI prostate localization using different MRI modality scans. Proceedings of the 2020 IEEE 8th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE), Vilnius, Lithuania.10.1109/AIEEE51419.2021.9435797 ↗
2. Weakly supervised deep learning for whole slide lung cancer image analysis10.1109/TCYB.2019.2935141 ↗
3. Detection of glaucoma using retinal fundus images: A comprehensive review10.3934/mbe.2021106 ↗
4. Glaucoma assessment from color fundus images using convolutional neural network10.1002/ima.22494 ↗
5. Amyar, A., Modzelewski, R., Li, H., and Ruan, S. (2020). Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: Classification and segmentation. Comput. Biol. Med., 126.10.1016/j.compbiomed.2020.104037 ↗
6. A deep feature learning model for pneumonia detection applying a combination of mRMR feature selection and machine learning models10.1016/j.irbm.2019.10.006 ↗
7. Travel Direction Recommendation Model Based on Photos of User Social Network Profile10.1109/ACCESS.2023.3260103 ↗
8. Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., and Wei, X. (2022). YOLOv6: A single-stage object detection framework for industrial applications. arXiv.
9. Intelligent small object detection for digital twin in smart manufacturing with industrial cyber-physical systems10.1109/TII.2021.3061419 ↗
10. Li, C., Wang, R., Li, J., and Fei, L. (2020). Recent Trends in Intelligent Computing, Communication and Devices: Proceedings of ICCD 2018, Springer.
11. YOLO-face: A real-time face detector10.1007/s00371-020-01831-7 ↗
12. Ye, X., Liu, Y., Zhang, D., Hu, X., He, Z., and Chen, Y. (2023). Rapid and Accurate Crayfish Sorting by Size and Maturity Based on Improved YOLOv5. Appl. Sci., 13.10.3390/app13158619 ↗
13. CA-YOLOv5: Detection model for healthy and diseased silkworms in mixed conditions based on improved YOLOv5
14. Hui, Y., You, S., Hu, X., Yang, P., and Zhao, J. (2024). SEB-YOLO: An Improved YOLOv5 Model for Remote Sensing Small Target Detection. Sensors, 24.10.3390/s24072193 ↗
15. Greenhouse tomato detection and pose classification algorithm based on improved YOLOv510.1016/j.compag.2023.108519 ↗
16. Real-time object detection method based on YOLOv5 and efficient mobile network10.1007/s11554-024-01433-9 ↗
17. Reddy, B.K., Bano, S., Reddy, G.G., Kommineni, R., and Reddy, P.Y. (2021, January 20–22). Convolutional network based animal recognition using YOLO and Darknet. Proceedings of the 2021 6th International Conference on Inventive Computation Technologies (ICICT), Coimbatore, India.10.1109/ICICT50816.2021.9358620 ↗
18. Deep convolutional neural network for enhancing traffic sign recognition developed on Yolo V410.1007/s11042-022-12962-5 ↗
19. Hameed, K., Chai, D., and Rassau, A. (2020). A sample weight and adaboost cnn-based coarse to fine classification of fruit and vegetables at a supermarket self-checkout. Appl. Sci., 10.10.3390/app10238667 ↗
20. (2024, January 13). Construction Details Dataset. Available online: https://app.box.com/s/j420ld0wo89hvh6np1rc3z9t1e65yg2k.
21. Kwon, H.J., Kim, H.G., and Lee, S.H. (2021). Pill detection model for medicine inspection based on deep learning. Chemosensors, 10.10.3390/chemosensors10010004 ↗
22. Mask R-CNN10.1109/TPAMI.2018.2844175 ↗
23. Redmon, J., and Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv.
24. Tan, L., Huangfu, T., Wu, L., and Chen, W. (2021). Comparison of RetinaNet, SSD, and YOLO v3 for real-time pill identification. BMC Med. Inform. Decis. Mak., 21.10.1186/s12911-021-01691-8 ↗
25. Ou, Y.Y., Tsai, A.C., Wang, J.F., and Lin, J. (2018, January 23–26). Automatic drug pills detection based on convolution neural network. Proceedings of the 2018 International Conference on Orange Technologies (ICOT), Nusa Dua, Indonesia.10.1109/ICOT.2018.8705849 ↗
26. Chollet, F. (2017, January 21–26). Xception: Deep learning with depthwise separable convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.195 ↗
27. Automatic drug pills detection based on enhanced feature pyramid network and convolution neural networks10.1049/iet-cvi.2019.0171 ↗
28. A robust approach for industrial small-object detection using an improved faster regional convolutional neural network10.1038/s41598-021-02805-y ↗
29. Yildiz, E., and Wörgötter, F. (2019, January 26–29). DCNN-based screw detection for automated disassembly processes. Proceedings of the 2019 15th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), Sorrento, Italy.10.1109/SITIS.2019.00040 ↗
30. Vision-based screw head detection for automated disassembly for remanufacturing10.1016/j.procir.2022.02.001 ↗
31. A review of object detection based on deep learning10.1007/s11042-020-08976-6 ↗
32. Zou, X. (2019, January 10–11). A review of object detection techniques. Proceedings of the 2019 International Conference on Smart Grid and Electrical Automation (ICSGEA), Xiangtan, China.10.1109/ICSGEA.2019.00065 ↗
33. Li, K., and Cao, L. (2020, January 15–17). A review of object detection techniques. Proceedings of the 2020 5th International Conference on Electromechanical Control Technology and Transportation (ICECTT), Nanchang, China.10.1109/ICECTT50890.2020.00091 ↗
34. Terven, J., and Cordova-Esparza, D. (2023). A comprehensive review of YOLO: From YOLOv1 to YOLOv8 and beyond. arXiv.
35. Zhao, Y., Shi, Y., and Wang, Z. (2020, January 5–7). The improved YOLOV5 algorithm and its application in small target detection. Proceedings of the International Conference on Intelligent Robotics and Applications, Kuala Lumpur, Malaysia.
36. Investigation of YOLOv5 Efficiency in iPhone Supported Systems
37. Kvietkauskas, T., and Stefanovič, P. (2023). Influence of Training Parameters on Real-Time Similar Object Detection Using YOLOv5s. Appl. Sci., 13.10.3390/app13063761 ↗
38. Optimizing the hyperparameter tuning of YOLOv5 for underwater detection10.1109/ACCESS.2022.3174583 ↗
39. Mantau, A.J., Widayat, I.W., Adhitya, Y., Prakosa, S.W., Leu, J.S., and Köppen, M. (2022, January 27–30). A GA-Based Learning Strategy Applied to YOLOv5 for Human Object Detection in UAV Surveillance System. Proceedings of the 2022 IEEE 17th International Conference on Control & Automation (ICCA), Naples, Italy.10.1109/ICCA54724.2022.9831954 ↗
40. Jocher, G., Chaurasia, A., Stoken, A., Borovec, J., Kwon, Y., Michael, K., Xie, T., Fang, J. (2021). ultralytics/yolov5: V7.0—YOLOv5 SOTA Realtime Instance Segmentation, Zenodo.
41. Huang, Q., Zhou, Y., Yang, T., Yang, K., Cao, L., and Xia, Y. (2023). A Lightweight Transfer Learning Model with Pruned and Distilled YOLOv5s to Identify Arc Magnet Surface Defects. Appl. Sci., 13.10.3390/app13042078 ↗
42. Ultralytics (2024, January 13). Hyperparameter Tuning. Ultralytics YOLOv8 Docs. 3 March 2024. Available online: https://docs.ultralytics.com/guides/hyperparameter-tuning.
43. Ultralytics (2024, January 24). “Train”. Ultralytics YOLOv8 Docs. 30 March 2024. Available online: https://docs.ultralytics.com/modes/train/#train-settings.
44. Ruman (2024, January 24). YOLO Data Augmentation Explained–Ruman–Medium. Medium. 4 June 2023. Available online: https://rumn.medium.com/yolo-data-augmentation-explained-turbocharge-your-object-detection-model-94c33278303a.