An Improved YOLOv5s Method for Light Guide Plate Defect Detection
DOI:
https://doi.org/10.54691/fse.v2i6.966Keywords:
Light Guide Plate Defect Detection, Improve YOLOv5s, Aattention MechanismAbstract
Aiming at the problems of over 100 million pixels in a single image of a large-size light guide plate, only a dozen pixels in the defect size, uneven brightness, and slow detection speed, this paper designs an improved YOLOv5s method for light guide plate defect detection. YOLOv5s is used as the basic model, and the attention mechanism CBAM is added to the backbone feature extraction network (backbone) part; PAN-Net is replaced with Bi-FPN structure in the feature fusion network (Neck) part, and a small target detection layer is added at the same time; YOLOv5s Obtain higher accuracy and improve algorithm robustness. Finally, a comparative analysis is performed based on the self-built dataset LGPDD. The experimental results show that compared with the original YOLOv5s model, the average accuracy mAP is increased by 1.4%; the model detection speed can meet the production requirements, and has a good application prospect in industrial deployment.
Downloads
References
Li Y, Li J. An end-to-end defect detection method for mobile phone light guide plate via multitask learning[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-13.
Hong L, Wu X, Zhou D, et al. Effective Defect Detection Method Based on Bilinear Texture Features for LGPs[J]. IEEE Access, 2021, 9: 147958-147966.
Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2014: 580-587.
Ren S, He K, Girshick R, et al. Faster r-cnn: Towards real-time object detection with region proposal networks[J]. Advances in neural information processing systems, 2015, 28.
Redmon J, Farhadi A. Yolov3: An incremental improvement[J]. arXiv preprint arXiv:1804.02767, 2018.
Bochkovskiy A, Wang C Y, Liao H Y M. Yolov4: Optimal speed and accuracy of object detection[J]. arXiv preprint arXiv:2004.10934, 2020.
Liu W, Anguelov D, Erhan D, et al. Ssd: Single shot multibox detector[C]//European conference on computer vision. Springer, Cham, 2016: 21-37.
Yao J, Li J. AYOLOv3-Tiny: An improved convolutional neural network architecture for real-time defect detection of PAD light guide plates[J]. Computers in Industry, 2022, 136: 103588.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
