Traffic safety helmet wear detection based on improved YOLOv5 network

Abstract

Aiming at the problem that the current traffic safety helmet detection model can’t balance the accuracy of detection with the size of the model and the poor generalization of the model, a method based on improving you only look once version 5 (YOLOv5) is proposed. By incorporating the lightweight GhostNet module into the YOLOv5 backbone network, we effectively reduce the model size. The addition of the receptive fields block (RFB) module enhances feature extraction and improves the feature acquisition capability of the lightweight model. Subsequently, the high-performance lightweight convolution, GSConv, is integrated into the neck structure for further model size compression. Moreover, the baseline model’s loss function is substituted with efficient insertion over union (EloU), accelerating network convergence and enhancing detection precision. Experimental results corroborate the effectiveness of this improved algorithm in real-world traffic scenarios.

References

[1] Espinosa J E, Velastin S A, Branch J W. Detection of motorcycles in urban traffic using video analysis: a review. IEEE transactions on intelligent transportation systems, 2020, 22(10): 6115-6130 J]

[2] Sugiarto R, Susanto E K, Kristian Y. Helmet usage detection on motorcyclist using deep residual learning. 2021 3rd East Indonesia Conference on Computer and Information Technology, April 9–11, 2021, Surabaya, Indonesia, 2021 New York IEEE 194-198 [C]

[3] He K, Zhang X, Ren S, et al.. Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 27–30, 2016, Las Vegas, NV, USA, 2016 New York IEEE 770-778 [C]

[4] Mistry J, Misraa A K, Agarwal M, et al.. An automatic detection of helmeted and non-helmeted motorcyclist with license plate extraction using convolutional neural network. 2017 7th International Conference on Image Processing Theory, Tools and Applications, November 28–December 1, 2017, Montreal, QC, Canada, 2017 New York IEEE 1-6 [C]

[5] Lin T Y, Maire M, Belongie S, et al.. Microsoft coco: common objects in context. Proceedings of the European Conference on Computer Vision (ECCV), September 6–12, 2014, Zurich, Switzerland, 2014 Heidelberg Springer International Publishing 740-755 [C]

[6] Dasgupta M, Bandyopadhyay O, Chatterji S. Automated helmet detection for multiple motorcycle riders using CNN. 2019 IEEE Conference on Information and Communication Technology, December 6–8, 2019, Allahabad, India, 2019 New York IEEE 1-4 [C]

[7] Li C H, Huang D. Detecting helmets on motorcyclists by deep neural networks with a dual-detection scheme. 28th International Conference on Neural Information Processing, December 8–12, 2021, Sanur, Bali, Indonesia, 2021 Heidelberg Springer International Publishing 417-427 [C]

[8] Li Z, Xie W, Zhang L, et al.. Toward efficient safety helmet detection based on YOLOv5 with hierarchical positive sample selection and box density filtering. IEEE transactions on instrumentation and measurement, 2022, 71: 1-14 J]

[9] Liu Y, Shi G, Li Y, et al.. M-YOLO: traffic sign detection algorithm applicable to complex scenarios. Symmetry, 2022, 14(5): 952 J]

[10] Charran R S, Dubey R K. Two-wheeler vehicle traffic violations detection and automated ticketing for Indian road scenario. IEEE transactions on intelligent transportation systems, 2022, 23(11): 22002-22007 J]

[11] Farid A, Hussain F, Khan K, et al.. A fast and accurate real-time vehicle detection method using deep learning for unconstrained environments. Applied sciences, 2023, 13(5): 3059 J]

[12] Cheng R, He X, Zheng Z, et al.. Multi-scale safety helmet detection based on SAS-YOLOv3-tiny. Applied sciences, 2021, 11(8): 3652 J]

[13] Liu S, Kong W, Chen X, et al.. Multi-scale ship detection algorithm based on a lightweight neural network for spaceborne SAR images. Remote sensing, 2022, 14(5): 1149 J]

[14] Sandler M, Howard A, Zhu M, et al.. Mobile-netv2: inverted residuals and linear bottle-necks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 18–23, 2018, Salt Lake City, UT, USA, 2018 New York IEEE 4510-4520 [C]

[15] Li S, Fu X, Dong J. Improved ship detection algorithm based on YOLOX for SAR outline enhancement image. Remote sensing, 2022, 14(16): 4070 J]

[16] GE Z, LIU S, WANG F, et al. YOLOX: exceeding YOLO series in 2021[EB/OL]. (2021-07-18) [2023-5-26]. https://arxiv.org/abs/2107.08430.

[17] Han K, Wang Y, Tian Q, et al.. Ghostnet: more features from cheap operations. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 13–19, 2020, Seattle, WA, USA, 2020 New York IEEE 1580-1589 [C]

[18] Xiao B, Yan C. A lightweight global awareness deep network model for flame and smoke detection. Optoelectronics letters, 2023, 19(10): 614-622 J]

[19] TARG S, ALMEIDA D, LYMAN K. Resnet in Resnet: generalizing residual architectures[EB/OL]. (2016-03-25) [2023-5-26]. https://arxiv.org/abs/1603.08029.

[20] Liu X Y, Guo C Y, Gong Z H, et al.. Object detection in remote sensing image with improved RFB net. Journal of geomatics science and technology, 2019, 36(2): 179-184 [J]

RIGHTS & PERMISSIONS

Tianjin University of Technology