Skip to main content

Advertisement

SpringerLink
Log in

Recognition of abnormal human behavior in dual-channel convolutional 3D construction site based on deep learning

  • S.I.: AI based Techniques and Applications for Intelligent IoT Systems
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Human behavior recognition has widespread applications in real life, especially in the infrastructure field. Modern engineering projects involve many construction units and engineering documents, many personnel, large scales, and large amounts of information. The traditional engineering management information system also has some defects. To strictly control the construction order and work efficiency of the infrastructure site, deal with emergencies in a timely manner, and reduce the burden on staff, we design a human abnormal behavior recognition system based on deep learning and dual-channel C3D. Our main aim is to conduct an in-depth study of human behavior recognition and apply it to the safety management of existing residential construction sites. In the methods section, we first introduce the convolutional neural network in deep learning, explain the principle of neural networks and two-channel convolution, and use the C3D model as the recognition model of the algorithm. The improved model combines the convolutional neural network to obtain our improved model. In the experimental section, the working process, experimental environment, and objects of our designed human abnormal behavior recognition system are classified. The analysis comprehensively analyzes the influence of system identification, the recall rate and accuracy rate of different algorithms, the misrecognition rate of different angles, dual-channel fusion, and the influence of duration on the prediction results. We chose three different angles for behavior recognition, and the results show that from a specific angle, our anomaly recognition rate is 98.01%; from a lateral point of view, our anomaly recognition rate is 97.27%; and from a top-down point of view, the recognition rate is 95.68%. Under three different video surveillance angles, the anomaly recognition rate of our proposed dual-channel 3D convolution algorithm reaches over 95%.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13

Similar content being viewed by others

Data availability

The data generated and analyzed during the current study are available from the corresponding author on reasonable request.

References

  1. Bin G, Yingying C, Nic L et al (2018) Behavior recognition based on Wi-Fi CSI: part 1[J]. IEEE Commun Mag 56(5):108–108

    Article  Google Scholar 

  2. Siamak, YH et al. (2017) A Survey on Behavior recognition using WiFi channel state information. IEEE Commun Mag 55(10):98–104

  3. Jingyi Qu, Zhu Wei Wu, Renbiao. (2017) Image classification for dual-channel neural networks based on attenuation factor. Syst Eng Electronics 39(6):1391–1399 ((In chinese))

    Google Scholar 

  4. Liling Z, Lan L, Xinpeng Z et al (2017) Driver behavior recognition based on support vector machine. J Transp Syst Eng Inf Technol 17(1):91–97 ((In chinese))

    Google Scholar 

  5. Lentzas A, Vrakas D (2020) Non-intrusive human activity recognition and abnormal behavior detection on elderly people: a review. Artif Intell Rev 53(3):1975–2021

    Article  Google Scholar 

  6. Zhiyong H, Shuaiyu D et al. (2019) Research on glioma magnetic resonance imaging segmentation based on dual-channel three-dimensional densely connected network. J Biomed Eng 36(5):763–768

  7. Litjens G, Kooi T, Bejnordi BE et al. (2017) A survey on deep learning in medical image analysis. Med Image Anal 42(9):60–88

  8. Xing H, Guigang Z, Shang Ma (2016) Deep learning. Int J Semantic Comput 10(03):417–439

    Article  Google Scholar 

  9. Dixon MF, Polson NG et al (2019) Deep learning for spatio-temporal modeling: Dynamic traffic flows and high frequency trading. Appl Stoch Model Bus Ind 35(3):788–807

    Article  MathSciNet  Google Scholar 

  10. Sirinukunwattana K, Shan E, Tsang YW et al (2016) Locality sensitive deep learning for detection and classification of nuclei in routine colon cancer histology images. IEEE Trans Med Imag 35(5):1196–1206

    Article  Google Scholar 

  11. Yushi C, Zhouhan L, Zhao X et al (2017) Deep learning-based classification of hyperspectral data. IEEE J Selected Top Appl Earth Observ Remote Sens 7(6):2094–2107

    Google Scholar 

  12. Park S, Ji M, Chun J (2018) 2D Human pose estimation based on object detection using RGB-D information. KSII Trans Internet Inf Syst 12(2):800–816

    Google Scholar 

  13. Qing Yu, Peijing J, Yaoguo W (2020) Research on first aid measures based on convolutional neural network recognition human actions. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 32(11):1385–1387

    Google Scholar 

  14. Jian L, Jin T, Peng Z et al (2016) Abnormal behavior detection for elderly based on 3D structure light sensor. Opt Technique 35(5):1271–1275 ((In chinese))

    Google Scholar 

  15. Basavaraj GM, Kusagur A (2018) Optical flow approach followed by SVM classification model to recognize abnormal behavior of a crowd. Int J Eng Technol 7(3):156–159

    Google Scholar 

  16. Bernhardt J, Büttner K, Scharf C et al (2015) Dual channel imaging of two-dimensional electropherograms in Bacillus subtilis. Electrophoresis 20(11):2225–2240

    Article  Google Scholar 

  17. Kılıçman A, Algabishi GO et al. (2015) On Classifications of non-constant linear second order Pdes and double convolution product. Far East J Math Sci 97(8):1031–1042

  18. Durdyev S, Ismail S (2016) On-site construction productivity in Malaysian infrastructure projects. Struct Surv 34(4–5):446–462

    Article  Google Scholar 

  19. Tempa C (2020) Geotechnical site investigation for infrastructure development project at Jigmeling Industrial Estate in Sarpang, Bhutan. Int J Eng Technol IJET 11(5):143–149

    Google Scholar 

  20. Kotwica K, Suffner H, Andras A (2020) Trenchless execution of microtunnels for municipal infrastructure in urbanized site—conception of a new solution. Manage Syst Prod Eng 28(4):276–282

    Google Scholar 

  21. Armitage J (2018) Design standardisation is the key to adopting off-site manufacturing in infrastructure. Proc Inst Civil Eng Civil Eng 171(1):13–13

    Google Scholar 

  22. Vereshchaka N (2020) Optimization of infrastructure project product parameters. Innov Technol Sci Solutions Ind 4(14):31–39

  23. Orlovskaya VI, Trifonov AG (2018) Dose assessment for typical NPP-2006 staff for design basis accident taking into account site infrastructure. Proc Nat Acad Sci Belarus Phys Tech Ser 63(1):111–118

    Google Scholar 

  24. Manju D (2018) A survey on human activity prediction techniques. Int J Adv Technol Eng Explor 5(47):400–406

    Article  Google Scholar 

  25. Huaqian L, Xiujuan Z, Yan W et al. (2021) A Gaussian mixture-hidden Markov model of human visual behavior. J Biomed Eng 38(3):512–519 (In chinese)

  26. Cui Jie Hu, Haidong CX et al (2018) Moving target detection for multi-beam forward-looking sonar based on frame-difference method. Chinese J Sci Inst 39(2):169–176

    Google Scholar 

  27. Huilin Y, Jie W, Jia L et al (2021) A memory-attention hierarchical model for driving-behavior recognition and motion prediction. Int J Automot Technol 22(4):895–908

    Article  Google Scholar 

Download references

Funding

This work was supported by the National Natural Science Foundation of China under Grant 61902435. We are grateful for resources from the High Performance Computing Center of Central South University.

Author information

Authors and Affiliations

  1. School of Computer Science and Engineering, Central South University, Changsha, 410083, Hunan, China

    Lingzi Jiang, Beiji Zou, Shu Liu, Min Wang & Enquan Huang

  2. Hunan Engineering Research Center of Machine Vision and Intelligent Medicine, Changsha, 410083, Hunan, China

    Lingzi Jiang, Beiji Zou, Shu Liu, Min Wang & Enquan Huang

  3. School of Computer Science, Hunan First Normal University, Changsha, 410002, Hunan, China

    Wenjun Yang

Authors
  1. Lingzi Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Beiji Zou
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Shu Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Wenjun Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Min Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Enquan Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shu Liu.

Ethics declarations

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jiang, L., Zou, B., Liu, S. et al. Recognition of abnormal human behavior in dual-channel convolutional 3D construction site based on deep learning. Neural Comput & Applic 35, 8733–8745 (2023). https://doi.org/10.1007/s00521-022-07881-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-022-07881-3

Keywords

Search

Navigation