Identification and application of unsafe behaviors of subway construction workers based on deep learning

doi:10.16265/j.cnki.issn1003-3033.2023.01.0874

Abstract

Abstract:

In order to effectively identify unsafe behaviors of subway construction workers, an unsafe behavior recognition method combining behavior and identity recognition was proposed based on deep learning and computer vision technology. Firstly, it is optimized the existing Faster R-CNN algorithm, and the ECA attention module is introduced to improve the accuracy of behavior recognition. Secondly, the face recognition method based on the face super-resolution algorithm is combined with behavior recognition to improve the pixel level of the image and accurately output the relevant information of the perpetrators of unsafe behavior. Thirdly, behavior recognition is performed concurrently with face recognition, and the results are streamed back to the database for the final output of the worker's unsafe behavior report. Finally, four unsafe behaviors from a subway construction project were selected for the empirical application of the recognition method. The research shows that the method can be effectively applied in subway construction scenarios, and the accuracy of both behavior recognition and face recognition reached above 0.85, which has high accuracy.

Key words: subway construction, unsafe behavior recognition, construction workers, faster region-based convolutional neural networks(Faster R-CNN), face recognition

FAN Bingqian, DONG Bingyu, WANG Biao, LI Ming, WU Song, TONG Ruipeng. Identification and application of unsafe behaviors of subway construction workers based on deep learning[J]. China Safety Science Journal, 2023, 33(1): 41-47.

Figures/Tables 6

Fig.1

Fig.2

Fig.3

Fig.4

Tab.1

Tab.2

References 18

[1]	FANG Weili, LOVE P E D, LUO Hanbin, et al. Computer vision for behaviour-based safety in construction: a review and future directions[J]. Advanced Engineering Informatics, 2020, 43:DOI: 10.1016/j.aei.2019.100980. doi: 10.1016/j.aei.2019.100980
[2]	RECARTE M Á, PÉREZ E, CONCHILLO Á, et al. Mental workload and visual impairment: differences between pupil, blink, and subjective rating[J]. The Spanish Journal of Psychology, 2008, 11(2):374-385. doi: 10.1017/S1138741600004406
[3]	张巧丽, 赵地, 迟学斌. 基于深度学习的医学影像诊断综述[J]. 计算机科学, 2017, 44(增2):1-7.
	ZHANG Qiaoli, ZHAO Di, CHI Xuebin. Review for deep learning based on medical imaging diagnosis[J]. Computer Science, 2017, 44(S2):1-7.
[4]	TIAN Hongkun, WANG Tianhai, LIU Yadong, et al. Computer vision technology in agricultural automation-a review[J]. Information Processing in Agriculture, 2019, 7(1):1-19. doi: 10.1016/j.inpa.2019.09.006
[5]	佟瑞鹏, 张艳伟. 人工智能技术在矿工不安全行为识别中的融合应用[J]. 中国安全科学学报, 2019, 29(1):7-12. doi: 10.16265/j.cnki.issn1003-3033.2019.01.002
	TONG Ruipeng, ZHANG Yanwei. Integration between artificial intelligence technologies for miners' unsafe behavior identification[J]. China Safety Science Journal, 2019, 29(1):7-12. doi: 10.16265/j.cnki.issn1003-3033.2019.01.002
[6]	YU Yantao, GUO Hongling, DING Qinghua, et al. An experimental study of real-time identification of construction workers' unsafe behaviors[J]. Automation in Construction, 2017, 82:193-206. doi: 10.1016/j.autcon.2017.05.002
[7]	温廷新, 王贵通, 孔祥博, 等. 基于迁移学习与残差网络的矿工不安全行为识别[J]. 中国安全科学学报, 2020, 30(3):41-46. doi: 10.16265/j.cnki.issn1003-3033.2020.03.007
	WEN Tingxin, WANG Guitong, KONG Xiangbo, et al. Identification of miners' unsafe behaviors based on transfer learning and residual network[J]. China Safety Science Journal, 2020, 30(3):41-46. doi: 10.16265/j.cnki.issn1003-3033.2020.03.007
[8]	FANG Weili, DING Lieyun, LUO Hanbin, et al. Falls from heights: a computer vision-based approach for safety harness detection[J]. Automation in Construction, 2018, 91:53-61. doi: 10.1016/j.autcon.2018.02.018
[9]	SEO J O, HAN S U, LEE S H, et al. Computer vision techniques for construction safety and health monitoring[J]. Advanced Engineering Informatics, 2015, 29(2):239-251. doi: 10.1016/j.aei.2015.02.001
[10]	FANG Qi, LI Heng, LUO Xiaochun, et al. A deep learning-based method for detecting non-certified work on construction sites[J]. Advanced Engineering Informatics, 2018, 35:56-68. doi: 10.1016/j.aei.2018.01.001
[11]	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]. Advances in Neural Information Processing Systems, 2012:1097-1105.
[12]	UIJLINGS J R R, VAN DE SANDE K E A, GEVERS T, et al. Selective search for object recognition[J]. International Journal of Computer Vision, 2013, 104(2):154-171. doi: 10.1007/s11263-013-0620-5
[13]	GIRSHICK R. Fast R-CNN[C]. Proceedings of the IEEE International Conference on Computer Vision, 2015:1440-1448.
[14]	REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards teal-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149. doi: 10.1109/TPAMI.2016.2577031 pmid: 27295650
[15]	SEVAKULA R K, SINGH V, VERMA N K, et al. Transfer learning for molecular cancer classification using deep neural networks[J]. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 2018, 16(6):2089-2100. doi: 10.1109/TCBB.2018.2822803
[16]	MA Cheng, JIANG Zhenyu, RAO Yaogming, et al. Deep face super-resolution with iterative collaboration between attentive recovery and landmark estimation[C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020:5569-5578.
[17]	WANG Qilong, WU Banggu, ZHU Pengfei, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11 531-11 539.
[18]	JIANG Zhenyu. Deep iterative collaboration for face super-resolution[EB/OL]. [2022-05-25]. https://github.com/Maclory/Deep-Iterative-Collaboration.

标签类别	数量	比例%
安全帽	2 360	29.01
无安全帽	2 828	34.76
翻越护栏	988	12.14
安全带	719	8.84
无安全带	1 241	15.25
合计	8 136	100

各标签类别AP					MAP
佩戴安全帽	未佩戴安全帽	翻越护栏	系安全带	未系安全带	MAP
0.93	0.91	0.72	0.88	0.82	0.854