Abstract
In recent years, the demand for automatic crowd behavior analysis has surged, driven by the need to ensure public safety and minimize casualties during events of public and religious significance. However, effectively analyzing the nonlinearities present in real-world crowd images and videos remains a challenge. To address this, research proposes a novel approach leveraging deep learning (DL) architectures for the segmentation and classification of human crowd behavior. Our method begins by collecting input from surveillance videos capturing crowd activity, which is then processed to remove noise and extract the crowd scene. Subsequently, we employ an expectation–maximization-based ZFNet architecture for accurate video segmentation. The segmented video is then classified using transfer exponential Conjugate Gradient Neural Networks, enhancing the precision of crowd behavior characterization. Our method has been proven effective in experimental analysis on many human crowd datasets, with significant results of average mean precision (MAP) of 59%, the mean square error (MSE) of 61%, accuracy in the training of 95%, validation precision of 95%, and selectivity of 88%. The potential of DL-based methods to advance crowd behavior analysis for improved privacy and security is highlighted by this study.
Similar content being viewed by others
Data availability
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
Abbreviations
- [d0,…, d9] :
-
Local Descriptor
- [D0…, Dt-1] :
-
SSIM value
- I :
-
Incomplete
- g :
-
Dimensional
- z :
-
Component memberships
- \(1,{\pi }_{1},\dots ,{\pi }_{k}\) :
-
Mulr multinomial distribution
- \(Q\left(\Psi :{\widehat{\Psi }}^{n}\right)\) :
-
Conditional expectation
- v :
-
Observed data
- \({z}_{i}^{n}\) :
-
Posterior probability
- \({\Gamma }_{i}\) :
-
Kth component of the mixture
- \(\widehat{\psi }\) :
-
Parameter set
- \(\widehat{\Psi }(t+1)\) :
-
Fresh parameter
- \({\dot{\pi }}_{k}^{n+1)}\) :
-
New estimates
- \({z}_{n}^{n}\) :
-
Digamma function
- h × w :
-
Height, width
- d :
-
Depth
- \({x}_{ij}\) :
-
Input vector
- \({f}_{ks}\) :
-
Function
- \({y}_{ij}\) :
-
Vector output
- \({I}_{n}(g)\) :
-
Average of a function g
- \(g\left({{\varvec{x}}}_{i}\right)\) :
-
Independent random variables
- \({W}_{n}\) :
-
Weight matrix
- \({B}_{m(j)}\) :
-
Bias and m is the number of inputs
- \(\Delta (f*)\) :
-
Constant
- \(\underset{j}{{\text{inf}}} {\int }_{{\mathbb{R}}^{d}}\) :
-
Fourier transform extension
- W :
-
Output weights
- \({{\varvec{w}}}_{j}^{0}\) :
-
Random variables
- \((a1,\dots ,am)\) :
-
Coefficients
- U(k) :
-
Projected convergence rate
- V(k) :
-
Observed consensus error
- \({z}_{i}(k)\) :
-
Iterates
- z :
-
First term
- z* :
-
Expected optimization error
- \(R(k)\) :
-
SGD's effectiveness
- \({\eta }_{t}\) :
-
Rate of learning
- \({\mathbf{d}}_{\mathbf{t}}\) :
-
Sample drawn
- d :
-
Random sample
- \(\Omega\) :
-
Sample space
- \({h}_{t+1}-{h}_{t}\) :
-
Random variable
- \({\mathbf{w}}^{t}\) :
-
The separation between the existing solution
- \({\mathbf{w}}^{*}\) :
-
Ideal solution
- \({h}_{t}\) :
-
Random variable
References
Tyagi B, Nigam S, Singh R (2022) A review of deep learning techniques for crowd behaviour analysis. Arch Computat Methods Eng 29(7):5427–5455
Chaudhary D, Kumar S, Dhaka VS (2022) Video based human crowd analysis using machine learning: a survey. Comput Methods Biomech Biomed Eng: Imaging Vis 10(2):113–131
Bruno A, Ferjani M, Sabeur Z, Arbab-Zavar B, Cetinkaya D, Johnstone L, ... Benaouda D (2022) High-level feature extraction for crowd behaviour analysis: a computer vision approach. In Image Analysis and Processing. ICIAP 2022 Workshops: ICIAP International Workshops, Lecce, Italy, May 23–27, 2022, Revised Selected Papers, Part II (pp. 59–70). Springer International Publishing, Cham
Kong YX, Wu RJ, Zhang YC, Shi GY (2023) Utilizing statistical physics and machine learning to discover collective behaviour on temporal social networks. Inf Process Manage 60(2):103190
Farooq MU, Mohamad Saad MN, Saleh Y, Daud Khan S (2022) Deep learning approach for divergence behaviour detection at high density crowd. In International Conference on Artificial Intelligence for Smart Community: AISC 2020, 17–18 December, UniversitiTeknologi Petronas, Malaysia (pp. 875–888). Springer Nature Singapore, Singapore
Sharma V, Mir RN, Singh C (2023) Scale-aware CNN for crowd density estimation and crowd behaviour analysis. Comput Electr Eng 106:108569
Bahamid A, Mohd Ibrahim A (2022) A review on crowd analysis of evacuation and abnormality detection based on machine learning systems. Neural Comput Appl 34(24):21641–21655
Bhuiyan MR, Abdullah J, Hashim N, Al Farid F (2022) Video analytics using deep learning for crowd analysis: a review. Multimed Tools Appl 81(19):27895–27922
Matkovic F, Ivasic-Kos M, Ribaric S (2022) A new approach to dominant motion pattern recognition at the macroscopic crowd level. Eng Appl Artif Intell 116:105387
Hou H, Wang L (2022) Measuring dynamics in evacuation behaviour with deep learning. Entropy 24(2):198
Pattan P, Arjunagi S (2022) A human behaviour analysis model to track object behaviour in surveillance videos. Measurement: Sensors 24:100454
Abpeikar S, Kasmarik K, Garratt M, Hunjet R, Khan MM, Qiu H (2022) Automatic collective motion tuning using actor-critic deep reinforcement learning. Swarm Evol Comput 72:101085
Zhang D, Li W, Gong J, Huang L, Zhang G, Shen S, ... Ma H (2022) HDRLM3D: a deep reinforcement learning-based model with human-like perceptron and policy for crowd evacuation in 3D environments. ISPRS Int J Geo-Inform 11(4):255
Lu Y, Ruan X, Huang J (2022) Deep reinforcement learning based on social spatial-temporal graph convolution network for crowd navigation. Machines 10(8):703
Liu T, Zheng Q, Tian L (2022) Application of distributed probability model in sports based on deep learning: deep belief network (DL-DBN) algorithm for human behaviour analysis. Comput Intell Neurosci 2022
Ha D, Tang Y (2022) Collective intelligence for deep learning: a survey of recent developments. Collective Intell 1(1):26339137221114870
Liang Z, Li L, Wang L (2022) Crowd-oriented behaviour simulation: reinforcement learning framework embedded with emotion model. In Artificial Intelligence: Second CAAI International Conference, CICAI 2022, Beijing, China, August 27–28, 2022, Revised Selected Papers, Part III (pp. 195–207). Springer Nature Switzerland, Cham
Choi T, Pyenson B, Liebig J, Pavlic TP (2022) Beyond tracking: using deep learning to discover novel interactions in biological swarms. Artif Life Robot 27(2):393–400
Poon KH, Wong PKY, Cheng JC (2022) Long-time gap crowd prediction using time series deep learning models with two-dimensional single attribute inputs. Adv Eng Inform 51:101482
Tiwari RG, Yadav SK, Misra A, Sharma A (2022) Classification of swarm collective motion using machine learning. In Human-Centric Smart Computing: Proceedings of ICHCSC 2022. Springer Nature Singapore, Singapore, pp 173–181
Chakole PD, Satpute VR, Cheggoju N (2022) Crowd behaviour anomaly detection using correlation of optical flow magnitude. J Phys: Conf Ser 2273(1):012023 (IOP Publishing)
Guo B, Liu Y, Liu S, Yu Z, Zhou X (2022) CrowdHMT: crowd intelligence with the deep fusion of human, machine, and IoT. IEEE Internet Things J 9(24):24822–24842
Tripathi SK (2022) Design and development of some methods and models for crowd analysis using computer vision and deep learning approaches.
Lalit R, Purwar RK (2022) Crowd abnormality detection using optical flow and glcm-based texture features. J Inform Technol Res (JITR) 15(1):1–15
Pai AK, Chandrahasan P, Raghavendra U, Karunakar AK (2023) Motion pattern-based crowd scene classification using histogram of angular deviations of trajectories. Vis Comput 39(2):557–567
Bala B, Kadurka RS, Negasa G (2022) Recognizing unusual activity with the deep learning perspective in crowd segment. In: A Fusion of Artificial Intelligence and Internet of Things for Emerging Cyber Systems. Springer, Cham, pp 171–181
Vidhyalakshmi M, Ramesh S, Bharathi ML, Kshirsagar PR, Rajaram A, Tirth V (2023) A comparative recognition research on excretory organism in medical applications using neural networks. Multimed Tools Appl 1–18
Shafiq M, Tian Z, Bashir AK, Du X, Guizani M (2020) CorrAUC: A malicious bot-IoT traffic detection method in IoT network using machine-learning techniques. IEEE Internet Things J 8(5):3242–3254
Shafiq M, Tian Z, Bashir AK, Du X, Guizani M (2020) IoT malicious traffic identification using wrapper-based feature selection mechanisms. Comput Secur 94:101863
Shafiq M, Tian Z, Bashir AK, Jolfaei A, Yu X (2020) Data mining and machine learning methods for sustainable smart cities traffic classification: a survey. Sustain Cities Soc 60:102177
Singh D, Kaur M, Alanazi JM, AlZubi AA, Lee HN (2022) Efficient evolving deep ensemble medical image captioning network. IEEE J Biomed Health Inform 27(2):1016–1025
Raina R, Gondhi NK, Chaahat, Singh D, Kaur M, Lee HN (2023) A systematic review on acute leukemia detection using deep learning techniques. Arch Computat Methods Eng 30(1):251–270
Funding
No funding is involved in this work.
Author information
Authors and Affiliations
Contributions
All authors are contributed equally to this work.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
No participation of humans takes place in this implementation process.
Human and animal rights
No violation of Human and Animal Rights is involved.
Conflict of interest
Conflict of interest is not applicable in this work.
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 (e.g. a society or other partner) 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.
About this article
Cite this article
Garg, S., Sharma, S., Dhariwal, S. et al. Human crowd behaviour analysis based on video segmentation and classification using expectation–maximization with deep learning architectures. Multimed Tools Appl (2024). https://doi.org/10.1007/s11042-024-18630-0
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11042-024-18630-0