Based on Improved Yolov8 Safety Helmet Detection

Fushun Shan ； Jin Wang

《International Core Journal of Engineering》 9卷12期 (2023/12) Pp. 91-96

https://doi.org/10.6919/ICJE.202312_9(12).0014

摘要

In order to solve the problem that the original Yolov8 model often misses and mistake in detecting whether workers wear safety helmets, this paper proposes a safety helmet target detection algorithm based on improved Yolov8N. The attention modules of CBAM, SimAM and SA are added to the backbone of the original model to enhance the ability of the network to extract effective features of the target. Moreover, the weighted bidirectional feature pyramid BiFPN is introduced to enhance feature fusion and improve the accuracy of target detection. Experimental data show that using the improved model in the homemade dataset, the mAP of detecting safety helmets increased by 1.3% and the recall increased by 4%.

關鍵字

Yolov8 Model ； Safety Helmets ； Attention Module ； Feature Pyramid BiFPN

參考文獻

Wei L , Dragomir A , Dumitru E , et al. SSD: Single Shot MultiBox Detector[J]. ECCV 2016: 14th European Conference on Computer Vision, 2016: 21-37.

Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European conference on computer vision (ECCV), 2018:3-19.

Zhang Q L, Yang Y B. Sa-net: Shuffle attention for deep convolutional neural networks[C]//ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2021: 2235-2239.

Liqiong Yang, Liqiang Cai and Song GU. Safety helmet wearing behavior detection based on machine learning method[J]. China Safety Science and Technology,2019,15(10):152-157.

Zejia HAN, Qingkun XIAO, Liqi Zhang. Reflective Coat Detection Algorithm Based on Improved SSD Safety Helmet [J]. Automation and Instrumentation,2021,36(09):63-68.