Skip Navigation Links
Journal of Vibration Testing and System Dynamics

C. Steve Suh (editor), Pawel Olejnik (editor),

Xianguo Tuo (editor)

Pawel Olejnik (editor)

Lodz University of Technology, Poland


C. Steve Suh (editor)

Texas A&M University, USA


Xiangguo Tuo (editor)

Sichuan University of Science and Engineering, China


Detecting Smoking Behaviors in Public Places Based on T-Yolov4-tiny

Journal of Vibration Testing and System Dynamics 6(4) (2022) 361--371 | DOI:10.5890/JVTSD.2022.12.002

Ke-Yuan Tang$^{1,2}$, Chuan-Li Liu$^{2}$, Le-Cai Cai$^{2}$, Kui Cheng$^2$, Xing Liu$^{1, 2}$, Shao-Song Duan$^{1, 2}$

$^{1}$ School of Automation and Information Engineering, Sichuan University of Science & Engineering, Zigong 643000, China

$^{2}$ Yibin University, Yibin 644000, China

Download Full Text PDF



Smoking or passive smoking is not only harmful to health, but also easy to cause fires. However, there has been a lack of effective supervision of smoking behavior in public places. Most of the existing supervision methods rely on on-site patrols by supervisors, video surveillance or smoke alarms. These methods have problems such as low efficiency and low accuracy. In order to solve the problems, this paper proposes the T-Yolov4-tiny detection algorithm based on the Yolo lightweight network Yolov4-tiny. Specifically, multiple convolution layers are added to the original network to improve the CSPBlock Network structure; the 1$\mathrm{\times}$1 convolutional layers are used to reduce the amount of network calculations; the scale of the input image is increased to improve the ability of small target detection; the K-means clustering algorithm is utilized to optimize the anchor box size considering the actual target size in the data set, in order to improve the accuracy of the model. Then, a smoking behavior data set (named as Smoking-YBU) was collected by both web crawlers and manual collection. The experimental results show that, compared with the Yolov4-tiny algorithm, the mean average precision (mAP) of the T-Yolov4-tiny algorithm proposed in this paper increases by 12.69\% on the Smoking-YBU, and the detection speed can also meet real-time requirements.


This article is supposed by College Students' Innovative Entrepreneurial Training Plan Program, Sichuan University of Science \& Engineering (cx2020163).


  1. [1]  Li, Q., Hsia, J., and Yang, G. (2011), Prevalence of smoking in China in 2010, New England Journal of Medicine, 364(25), 2469-2470.
  2. [2]  \"{O}berg, M., Jaakkola, M.S., Woodward, A., Peruga, A., and Pr\"{u}ss-Ust\"{u}n, A. (2011), Worldwide burden of disease from exposure to second-hand smoke: a retrospective analysis of data from 192 countries, The Lancet, 377(9760), 139-146.
  3. [3]  Sacks, J.J. and Nelson, D.E. (1994), Smoking and injuries: an overview, Preventive Medicine, 23(4), 515-520.
  4. [4]  Stillman, F., Navas-Acien, A., Ma, J., Ma, S., Avila-Tang, E., Breysse, P., Yang, G., and Samet, J. (2007), Second-hand tobacco smoke in public places in urban and rural China, Tobacco Control, 16(4), 229-234.
  5. [5]  Hesketh, T., Ding, Q.-J., and Tomkins, A. (2001), Smoking among youths in China, American Journal of Public Health, 91(10), 1653-1655.
  6. [6]  Ballard, J.E., Koepsell, T.D., and Rivara, F. (1992), Association of smoking and alcohol drinking with residential fire injuries, American Journal of Epidemiology, 135(1), 26-34.
  7. [7]  Bien, T.L. and Lin, C.H. (2013), Detection and recognition of indoor smoking events, Fifth International Conference on Machine Vision (ICMV 2012): Algorithms, Pattern Recognition, and Basic Technologies, 8784, 878424.
  8. [8]  Wu, W.-C. and Chen, C.-Y. (2011), Detection system of smoking behavior based on face analysis, 2011 Fifth International Conference on Genetic and Evolutionary Computing, 184-187.
  9. [9]  Cui, J., Wang, L., Gu, T., Tao, X., and Lu, J. (2016), An audio-based hierarchical smoking behavior detection system based on a smart neckband platform, Proceedings of the 13th International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, New York, NY, USA, 190-199.
  10. [10]  Shoaib, M., Scholten, H., Havinga, P.J.M., and Incel, O.D. (2016), A hierarchical lazy smoking detection algorithm using smartwatch sensors, 2016 IEEE 18th International Conference on e-Health Networking, Applications and Services (Healthcom), 1-6.
  11. [11]  Zheng, X., Wang, J., Shangguan, L., Zhou, Z., and Liu, Y. (2016), Smokey: ubiquitous smoking detection with commercial WiFi infrastructures, IEEE INFOCOM 2016 - The 35th Annual IEEE International Conference on Computer Communications, 1-9.
  12. [12]  Senyurek, V.Y., Imtiaz, M.H., Belsare, P., Tiffany, S., and Sazonov, E. (2019), Smoking detection based on regularity analysis of hand to mouth gestures, Biomedical Signal Processing and Control, 51, 106-112.
  13. [13]  Yang, Z. and Yao, D. (2020), Fast TLAM: high-precision fine grain smoking behavior detection network, 2020 IEEE 3rd International Conference on Information Communication and Signal Processing (ICICSP), 183-188.
  14. [14]  Fu, L., Feng, Y., Wu, J., Liu, Z., Gao, F., Majeed, Y., Al-Mallahi, A., Zhang, Q., Li, R., and Cui, Y. (2020), Fast and accurate detection of kiwifruit in orchard using improved YOLOv3-tiny model, Precision Agric.
  15. [15]  Han, B.-G., Lee, J.-G., Lim, K.-T., and Choi, D.-H. (2020), Design of a scalable and fast YOLO for edge-computing devices, Sensors, 20(23), 6779.
  16. [16]  Bochkovskiy, A., Wang, C.-Y., and Liao, H.-Y.-M. (2020), YOLOv4: optimal speed and accuracy of object detection, arXiv preprint arXiv:2004.10934.
  17. [17]  Jiang, Z., Zhao, L., Li, S., and Jia, Y. (2020), Real-time object detection method based on improved YOLOv4-tiny, arXiv preprint arXiv:2011.04244.
  18. [18]  Byeon, Y.H. and Kwak, K.C. (2017) A performance comparison of pedestrian detection using faster RCNN and ACF, 2017 6th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI), 858-863.