Research on Detection Method of Daily Necessities based on YOLOv5s
DOI:
https://doi.org/10.54097/fcis.v5i1.11657Keywords:
YOLOv5, Object Detection, K-means Clustering, ECA-Net, BiFPNAbstract
A daily necessities detection algorithm based on improved YOLOv5 is proposed to address the issue of low detection accuracy in daily necessities. Firstly, replace the Euclidean distance of the original K-means clustering algorithm with a 1-IOU approach to detect anchor boxes that are more closely aligned with the target object. Secondly, in order to facilitate and quickly fuse multi-scale features in the network, BiFPN is used in the Neck layer to replace the original FPN+PAN structure. Then, in order to reduce the interference of the background environment on the algorithm, improve the performance and generalization ability of the model, the ECA-Net attention mechanism is introduced after BiFPN. Finally, the feasibility of the proposed method will be verified through detection experiments supported by the self-made dataset. The experimental structure shows that the improved method can improve the detection accuracy of daily necessities, with an accuracy increase of 3.30% compared to the original YOLOv5 algorithm.
Downloads
References
X F Miao, B L Liu, X Q Li, et al. Improvement of YOLOV5s Rail Crack Target Detection Algorithm [J/OL]. Computer Engineering and Applications: 1-11 [2023-09-02].
P Y Wang, L Mao. Research on small target detection algorithm for lychee fruit based on YOLOv5 [J]. Shanxi Electronic Technology, 2023 (04): 74-77H. Poor, An Introduction to Signal Detection and Estimation. New York: Springer-Verlag, 1985, ch. 4.
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.Columbus:IEEE Computer Society,2014:580.
REN S,HE K,GIRSHICK R,et al.Faster R-CNN:Towards Real-time Object Detection with Region Proposal Networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016,39(6):1137-1149.
Y Zhang, Z Y Gong, W W Wei. Traffic sign detection based on improved Faster R-CNN model [J]. Progress in Laser and Optoelectronics, 2020,57 (18): 173-181C. J. Kaufman, Rocky Mountain Research Lab., Boulder, CO, private communication, May 1995.
Y L Pei, H Luo, S H Zhang, et al. High speed rail fastener detection algorithm based on improved Faster R-CNN [J]. Journal of East China Jiaotong University, 2023,40 (01): 75-81M. Young, The Techincal Writers Handbook. Mill Valley, CA: University Science, 1989.
REDMON J, FARHADI A. YOLOv3: an incremental improvement [J]. arXiv e-prints,2018.
BOCHKOVSKIY A,WANG C Y,LIAO.YOLOv4:optimal speed and accuracy of object detection[R].2020.
X Y Niu, P G Mao, Y T Duan, et al. Research on Lightweight Improved Algorithm for Indoor Target Detection Based on YOLOv5s [J/OL]. Computer Engineering and Applications: 1-11 [2023-09-02].
G L Yang, J X Wang, Z L Nie. A real-time tomato recognition method based on improved YOLOv5s [J]. Jiangsu Agricultural Science, 2023,51 (15): 187-193.
C Zhang, Z Tian, J Song, et al. Construction worker hardhat-wearing detection based on an improved BiFPN[C]//2020 25th international conference on pattern recognition (ICPR). IEEE, 2021: 8600-8607.
Q Wang, B Wu, P Zhu, et al. ECA-Net: Efficient channel attention for deep convolutional neural networks[C]// Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020: 11534-11542.
R Zhou, M Li, S G Meng, et al. Research on Airport Runway Intrusion Warning Technology Based on YOLOv5 and Deep SORT [J/OL]. Electronic Measurement Technology: 1-6 [2023- 09-02].
Downloads
Published
Issue
Section
How to Cite
