Crowd Density Level Estimation and Anomaly Detection Using Multicolumn Multistage Bilinear Convolution Attention Network (MCMS-BCNN-Attention)
Abstract
:1. Introduction
Contribution of the Work
2. Related Works
3. Proposed Work
3.1. Dense Block
3.2. Bilinear Attention Feature Vector (Attention over BCNN Vector)
3.3. Proposed Model Architecture
3.4. Fully Connected Network (FCN)
4. Experiment Results and Discussion
4.1. Dataset Preparation
4.1.1. UCSD Dataset
4.1.2. Pets2009 Dataset
4.1.3. Umn Dataset
4.2. Model Training and Initialization of Parameters
4.3. Evaluation Metrics
4.4. Confusion Matrix
5. Results Analysis
5.1. Benchmark Datasets Analysis
5.2. Abnormal Event Detection and Classification
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Huang, G.; Liu, Z.; Van Der Maaten, L.; Weinberger, K.Q. Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2017, Honolulu, HI, USA, 21–26 July 2017; pp. 4700–4708. [Google Scholar]
- Tan, M.; Le, Q. Efficientnetv2: Smaller models and ster training. In Proceedings of the International Conference on Machine Learning, PMLR, Virtual Event, 18–24 July 2021; pp. 10096–10106. [Google Scholar]
- Zhang, Y.; Zhou, D.; Chen, S.; Gao, S.; Ma, Y. Single-image crowd counting via multi-column convolutional neural network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 589–597. [Google Scholar]
- Aldayri, A.; Albattah, W. Taxonomy of Anomaly Detection Techniques in Crowd Scenes. Sensors 2022, 22, 6080. [Google Scholar] [CrossRef] [PubMed]
- Wang, T.; Miao, Z.; Chen, Y.; Zhou, Y.; Shan, G.; Snoussi, H. Aed-net: An abnormal event detection network. Engineering 2019, 5, 930–939. [Google Scholar] [CrossRef]
- Biswas, S.; Babu, R.V. Anomaly detection via short local trajectories. Neurocomputing 2017, 242, 63–72. [Google Scholar] [CrossRef]
- Bera, A.; Kim, S.; Manocha, D. Realtime anomaly detection using trajectory-level crowd behavior learning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Las Vegas, NV, USA, 27–30 June 2016; pp. 50–57. [Google Scholar]
- Maiorano, F.; Petrosino, A. Granular trajectory based anomaly detection for surveillance. In Proceedings of the 2016 23rd International Conference on Pattern Recognition (ICPR), Cancun, Mexico, 4–8 December 2016; pp. 2066–2072. [Google Scholar]
- Biswas, S.; Babu, R.V. Short local trajectory based moving anomaly detection. In Proceedings of the 2014 Indian Conference on Computer Vision Graphics and Image Processing, Bangalore, India, 14–18 December 2014; pp. 1–8. [Google Scholar]
- Zhao, K.; Liu, B.; Li, W.; Yu, N.; Liu, Z. Anomaly detection and localization: A novel two-phase framework based on trajectory-level characteristics. In Proceedings of the 2018 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), San Diego, CA, USA, 23–27 July 2018; pp. 1–6. [Google Scholar]
- Zhang, X.; Ma, D.; Yu, H.; Huang, Y.; Howell, P.; Stevens, B. Scene perception guided crowd anomaly detection. Neurocomputing 2020, 414, 291–302. [Google Scholar] [CrossRef]
- Hao, Y.; Xu, Z.J.; Liu, Y.; Wang, J.; Fan, J.L. Effective crowd anomaly detection through spatio-temporal texture analysis. Int. J. Autom. Comput. 2019, 16, 27–39. [Google Scholar] [CrossRef] [Green Version]
- Li, N.; Chang, F. Video anomaly detection and localization via multivariate Gaussian fully convolution adversarial autoencoder. Neurocomputing 2019, 369, 92–105. [Google Scholar] [CrossRef]
- Li, X.; Li, W.; Liu, B.; Liu, Q.; Yu, N. Object-oriented anomaly detection in surveillance videos. In Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Calgary, AB, Canada, 15–20 April 2018; pp. 1907–1911. [Google Scholar]
- Si, C.; Chen, W.; Wang, W.; Wang, L.; Tan, T. An attention enhanced graph convolutional lstm network for skeleton-based action recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2019, Long Beach, CA, USA, 15–20 June 2019; pp. 1227–1236. [Google Scholar]
- Lloyd, K.; Rosin, P.L.; Marshall, D.; Moore, S.C. Detecting violent and abnormal crowd activity using temporal analysis of grey level co-occurrence matrix (GLCM) -based texture measures. Mach. Vis. Appl. 2017, 28, 361–371. [Google Scholar] [CrossRef] [Green Version]
- Hasan, M.; Choi, J.; Neumann, J.; Roy-Chowdhury, A.K.; Davis, L.S. Learning temporal regularity in video sequences. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2016, Las Vegas, NV, USA, 27–30 June 2016; pp. 733–742. [Google Scholar]
- Luo, W.; Liu, W.; Gao, S. Remembering history with convolutional lstm for anomaly detection. In Proceedings of the 2017 IEEE International Conference on Multimedia and Expo (ICME) 2017, Hong Kong, China, 10–14 July 2017; pp. 439–444. [Google Scholar]
- Gong, D.; Liu, L.; Le, V.; Saha, B.; Mansour, M.R.; Venkatesh, S.; Hengel, A.V. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision 2019, Seoul, Republic of Korea, 27 October–2 November 2019; pp. 1705–1714. [Google Scholar]
- Wu, C.; Shao, S.; Tunc, C.; Hariri, S. Video anomaly detection using pre-trained deep convolutional neural nets and context mining. In Proceedings of the 2020 IEEE/ACS 17th International Conference on Computer Systems and Applications (AICCSA), Antalya, Turkey, 2–5 November 2020; pp. 1–8. [Google Scholar]
- Huang, S.; Huang, D.; Zhou, X. Learning multimodal deep representations for crowd anomaly event detection. Math. Probl. Eng. 2018, 2018, 6323942. [Google Scholar] [CrossRef] [Green Version]
- Wang, T.; Qiao, M.; Zhu, A.; Shan, G.; Snoussi, H. Abnormal event detection via the analysis of multi-frame optical flow information. Front. Comput. Sci. 2020, 14, 304–313. [Google Scholar] [CrossRef]
- Singh, K.; Rajora, S.; Vishwakarma, D.K.; Tripathi, G.; Kumar, S.; Walia, G.S. Crowd anomaly detection using aggregation of ensembles of fine-tuned convnets. Neurocomputing 2020, 371, 188–198. [Google Scholar] [CrossRef]
- Sultani, W.; Chen, C.; Shah, M. Real-world anomaly detection in surveillance videos. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2018, Salt Lake City, UT, USA, 18–22 June 2018; pp. 6479–6488. [Google Scholar]
- Marsden, M.; McGuinness, K.; Little, S.; O’Connor, N.E. Resnetcrowd: A residual deep learning architecture for crowd counting, violent behaviour detection and crowd density level classification. In Proceedings of the 2017 14th IEEE international conference on advanced video and signal based surveillance (AVSS), Lecce, Italy, 29 August 2017–1 September 2017; pp. 1–7. [Google Scholar]
- Ratre, A. Taylor series based compressive approach and Firefly support vector neural network for tracking and anomaly detection in crowded videos. J. Eng. Res. 2019, 7, 115–137. [Google Scholar]
- Feng, Y.; Yuan, Y.; Lu, X. Learning deep event models for crowd anomaly detection. Neurocomputing 2017, 219, 548–556. [Google Scholar] [CrossRef]
- Fu, M.; Xu, P.; Li, X.; Liu, Q.; Ye, M.; Zhu, C. Fast crowd density estimation with convolutional neural networks. Eng. Appl. Artif. Intell. 2015, 43, 81–88. [Google Scholar] [CrossRef]
- A Comprehensive Hands-on Guide to Transfer Learning with Real-World Applications in Deep Learning. Available online: https://towardsdatascience.com/a-comprehensive-hands-on-guide-to-transfer-learning-with-real-world-applications-in-deep-learning-212bf3b2f27a (accessed on 1 November 2022).
- Lin, T.Y.; Roy Chowdhury, A.; Maji, S. Bilinear cnn models for fine-grained visual recognition. In Proceedings of the IEEE International Conference on Computer Vision 2015, Santiago, Chile, 7–13 December 2015; pp. 1449–1457. [Google Scholar]
- Kingma, D.P.; Ba, J. Adam: A method for stochastic optimization. arXiv 2014, arXiv:1412.6980. [Google Scholar]
- Paszke, A.; Gross, S.; Massa, F.; Lerer, A.; Bradbury, J.; Chanan, G.; Killeen, T.; Lin, Z.; Gimelshein, N.; Antiga, L.; et al. Pytorch: An imperative style, high-performance deep learning library. In Proceedings of the Advances in Neural Information Processing Systems 32 (NeurIPS 2019), Vancouver, BC, Canada, 8–14 December 2019; Neural Information Processing Systems Foundation, Inc. (NeurIPS): San Diego, CA, USA, 2019. Available online: https://proceedings.neurips.cc/paper/2019/file/bdbca288fee7f92f2bfa9 f7012727740-Paper.pdf (accessed on 2 December 2022).
- UCSD Anomaly Detection Dataset. Available online: http://www.svcl.ucsd.edu/projects/anomaly/dataset.html (accessed on 7 November 2022).
- PETS 2009 Benchmark Data. Available online: http://cs.binghamton.edu/mrldata/pets2009 (accessed on 7 November 2022).
- Monitoring Human Activity-Action Recognition. Available online: http://mha.cs.umn.edu/projrecognition.shtml (accessed on 7 November 2022).
- Alanazi, A.A.; Bilal, M. Crowd density estimation using novel feature descriptor. arXiv 2019, arXiv:1905.05891. [Google Scholar]
- Tripathy, S.K.; Srivastava, R. A real-time two-input stream multi-column multi-stage convolution neural network (TIS-MCMS-CNN) for efficient crowd congestion-level analysis. Multimed. Syst. 2020, 26, 585–605. [Google Scholar] [CrossRef]
- Shmueli, B.; Multi-Class Metrics Made Simple, Part III: The Kappa Score (Aka Cohen’s Kappa Coefficient). Medium. Towards Data Science. 2020. Available online: https://towardsdatascience.com/multi-class-metrics-made-simple-the-kappa-score-aka-cohens-kappa-coefficient-bdea137af09c (accessed on 2 December 2022).
- Chen, C.; Zhang, B.; Su, H.; Li, W.; Wang, L. Land-use scene classification using multi-scale completed local binary patterns. Signal Image Video Process. 2016, 10, 745–752. [Google Scholar] [CrossRef]
- Bradley, A.P. The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognit. 1997, 30, 1145–1159. [Google Scholar] [CrossRef] [Green Version]
- Wang, T.; Qiao, M.; Chen, Y.; Chen, J.; Zhu, A.; Snoussi, H. Video feature descriptor combining motion and appearance cues with length-invariant characteristics. Optik 2018, 157, 1143–1154. [Google Scholar] [CrossRef]
- Cong, Y.; Yuan, J.; Liu, J. Sparse reconstruction cost for abnormal event detection. In Proceedings of the Computer Vision and Pattern Recognition 2011, Colorado Springs, CO, USA, 20–25 June 2011; pp. 3449–3456. [Google Scholar]
- Wang, T.; Qiao, M.; Deng, A.; Zhou, Y.; Wang, H.; Lyu, Q.; Snoussi, H. Abnormal event detection based on analysis of movement information of video sequence. Optik 2018, 152, 50–60. [Google Scholar] [CrossRef]
- Susan, S.; Hanmandlu, M. Unsupervised detection of nonlinearity in motion using weighted average of non-extensive entropies. Signal Image Video Process. 2015, 9, 511–525. [Google Scholar] [CrossRef]
- Zhang, X.; Yang, S.; Tang, Y.Y.; Zhang, W. A thermodynamics-inspired feature for anomaly detection on crowd motions in surveillance videos Multimed. Tools Appl. 2020, 75, 8799–8826. [Google Scholar] [CrossRef]
- Kaltsa, V.; Briassouli, A.; Kompatsiaris, I.; Hadjileontiadis, L.J.; Strintzis, M.G. Swarm intelligence for detecting interesting events in crowded environments. IEEE Trans. Image Process. 2015, 24, 2153–2166. [Google Scholar] [CrossRef] [PubMed]
- Xu, Y.; Lu, L.; Xu, Z.; He, J.; Zhou, J.; Zhang, C. Dual-channel CNN for efficient abnormal behavior identification through crowd feature engineering. Mach. Vis. Appl. 2019, 30, 945–958. [Google Scholar] [CrossRef] [Green Version]
- Mu, H.; Sun, R.; Yuan, G.; Li, J.; Wang, M. Crowd behavior detection in videos using statistical physics. In Proceedings of the 2021 International Conference on Data Mining Workshops (ICDMW), Auckland, New Zealand, 7–10 December 2021; pp. 389–397. [Google Scholar]
- Ilyas, Z.; Aziz, Z.; Qasim, T.; Bhatti, N.; Hayat, M.F. A hybrid deep network based approach for crowd anomaly detection. Multimed. Tools Appl. 2021, 80, 24053–24067. [Google Scholar] [CrossRef]
- Du, Y. An anomaly detection method using deep convolution neural network for vision image of robot. Multimed. Tools Appl. 2020, 79, 9629–9642. [Google Scholar] [CrossRef]
- Singh, G.; Kapoor, R.; Khosla, A. Optical flow-based weighted magnitude and direction histograms for the detection of abnormal visual events using combined classifier. Int. J. Cogn. Inform. Nat. Intell. (IJCINI) 2021, 15, 12–30. [Google Scholar] [CrossRef]
- Xu, J.; Denman, S.; Fookes, C.; Sridharan, S. Unusual scene detection using distributed behaviour model and sparse representation. In Proceedings of the 2012 IEEE Ninth International Conference on Advanced Video and Signal-Based Surveillance, Beijing, China, 18–21 September 2012; pp. 48–53. [Google Scholar]
- Zhu, X.; Liu, J.; Wang, J.; Fu, W.; Lu, H. Weighted interaction force estimation for abnormality detection in crowd scenes. In Asian Conference on Computer Vision; Springer: Berlin/Heidelberg, Germany, 2012; pp. 507–518. [Google Scholar]
Input | 3 × 299 × 299 | ||
---|---|---|---|
Layer | Output | Filter Size | Neurons |
Convolution | 112 × 112 | 7 × 7 | 3 × 64 |
Pooling | 56 × 56 | 3 × 3 | 64 |
Dense Block | 56 × 56 | (1 × 1, 3 × 3) 6 | 64 × 256 |
Transition | (56 × 56, 28 × 28) | (1 × 1, 2 × 2) | 256 × 128 |
Dense Block | 28 × 28 | (1 × 1, 3 × 3) 12 | 128 × 512 |
Transition | (28 × 28, 14 × 14) | (1 × 1, 2 × 2) | 512 × 256 |
Dense Block | 14 × 14 | (1 × 1, 3 × 3) 24 | 256 × 1024 |
Transition | (14 × 14, 7 × 7) | (1 × 1, 2 × 2) | 1024 × 512 |
Dense Block | 7 × 7 | (1 × 1, 3 × 3) 16 | 512 × 1024 |
Dropout (0.5) | |||
Attention | 1024 | ||
FC | 512 |
Input Image | 3 × 299 × 299 (Channels × Width × Height) | |||
---|---|---|---|---|
Layer Name | Filter Size | Output Channels | Layer Name | Neurons |
Conv1 | 9 × 9 | 16 | Fusion | (40 × 1 × 1) |
Mp1 | 2 × 2 | FC1 | 1024 | |
Conv2 | 7 × 7 | 32 | FC | 512 |
Mp2 | 2 × 2 | |||
Conv3 | 7 × 7 | 16 | ||
Conv4 | 7 × 7 | 8 |
Input Image | 3 × 299 × 299 (Channels × Width × Height) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Layer Name | Branch 2 | Branch 3 | Layer Name | Neurons | ||||||
Layer Name | Filter Size | Output Channels | Layer Name | Filter Size | Output Channels | Layer Name | Filter Size | Output Channels | Fusion | (30 × 1 × 1) |
Conv1 | 9 × 9 | 16 | Conv1 | 7 × 7 | 20 | Conv1 | 5 × 5 | 24 | FC1 | 1024 |
Mp1 | 2 × 2 | Mp1 | 2 × 2 | Mp1 | 2 × 2 | FC | 512 | |||
Conv2 | 7 × 7 | 32 | Conv2 | 5 × 5 | 40 | Conv2 | 3 × 3 | 48 | ||
Mp2 | 2 × 2 | Mp2 | 2 × 2 | Mp2 | 2 × 2 | |||||
Conv3 | 7 × 7 | 16 | Conv3 | 5 × 5 | 20 | Conv3 | 3 × 3 | 24 | ||
Conv4 | 7 × 7 | 8 | Conv4 | 5 × 5 | 10 | Conv4 | 3 × 3 | 12 |
Dataset | Training Samples | Testing Samples | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
VL | L | M | H | VH | Total | VL | L | M | H | VH | Total | |
UCSD-Ped1 | 2093 | 4171 | 1043 | 614 | 119 | 8040 | 1394 | 2715 | 690 | 480 | 82 | 5360 |
UCSD-Ped2 | 2113 | 4137 | 1016 | 785 | 107 | 8158 | 1374 | 2749 | 717 | 502 | 98 | 5440 |
PETS2009 | 49 | 128 | 156 | 114 | 290 | 737 | 46 | 86 | 111 | 70 | 179 | 492 |
UMN-Plaza1 | 113 | 503 | 123 | 93 | 190 | 1022 | 79 | 335 | 76 | 66 | 126 | 682 |
UMN-Plaza2 | 197 | 570 | 100 | 189 | 143 | 1199 | 125 | 371 | 75 | 142 | 87 | 800 |
Predicted Class | ||
---|---|---|
Actual Class | Positive | Negative |
Positive | True positive (TP) | False negative (FN) |
Negative | false positive (FP) | True negative (TN) |
Dataset | Approach | Performance (%) | ||||
---|---|---|---|---|---|---|
Accuracy | Precision | Recall | F1 Score | Kappa | ||
UCSD Ped1 | CLBP [39] | 60.76 | 70.35 | 37.20 | 48.67 | 0.2238 |
MSCNN [28] | 97.48 | 98.70 | 96.22 | 97.45 | 0.9341 | |
densenet121 [1] | 94.32 | 97.20 | 91.27 | 94.15 | 0.8673 | |
Efficientnetv2 [2] | 97.03 | 98.55 | 95.46 | 96.98 | 0.9195 | |
Our (MCMS-BCNN-Attention+densenet121) | 98.28 | 99.31 | 97.24 | 98.26 | 0.9646 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 98.62 | 99.51 | 97.21 | 98.61 | 0.9576 | |
UCSD Ped2 | CLBP [39] | 62.19 | 73.80 | 37.78 | 49.98 | 0.2819 |
MSCNN [28] | 98.39 | 99.67 | 97.10 | 98.37 | 0.9834 | |
densenet121 [1] | 92.88 | 96.96 | 88.55 | 92.56 | 0.8656 | |
Efficientnetv2 [2] | 96.37 | 98.34 | 94.33 | 96.30 | 0.9231 | |
Our (MCMS-BCNN-Attention+densenet121) | 98.95 | 99.54 | 98.36 | 98.95 | 0.9794 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 98.08 | 99.23 | 96.91 | 98.05 | 0.9629 | |
PETS2009 | CLBP [39] | 86.80 | 95.37 | 77.33 | 85.42 | 0.8091 |
MSCNN [28] | 95.43 | 98.47 | 92.29 | 95.28 | 0.9222 | |
densenet121 [1] | 93.17 | 97.02 | 89.09 | 92.88 | 0.8581 | |
Efficientnetv2 [2] | 96.55 | 98.71 | 94.32 | 96.47 | 0.9327 | |
Our (MCMS-BCNN-Attention+densenet121) | 96.38 | 98.64 | 94.06 | 96.30 | 0.9273 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 96.97 | 98.81 | 95.08 | 96.91 | 0.9353 | |
UMN Plaza1 | CLBP [39] | 81.86 | 92.18 | 69.62 | 79.33 | 0.6857 |
MSCNN [28] | 97.79 | 99.24 | 96.31 | 97.76 | 0.9563 | |
densenet121 [1] | 96.85 | 98.78 | 94.86 | 96.78 | 0.9404 | |
Efficientnetv2 [2] | 98.50 | 99.41 | 97.57 | 98.48 | 0.9723 | |
Our (MCMS-BCNN-Attention + densenet121) | 98.50 | 99.33 | 97.67 | 98.49 | 0.9640 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 99.10 | 99.41 | 98.78 | 99.09 | 0.9642 | |
UMN Plaza2 | CLBP [39] | 60.36 | 70.18 | 36.01 | 47.60 | 0.2681 |
MSCNN [28] | 88.11 | 94.44 | 80.98 | 87.19 | 0.7841 | |
densenet121 [1] | 94.11 | 97.34 | 90.70 | 93.90 | 0.8843 | |
Efficientnetv2 [2] | 97.33 | 98.82 | 95.79 | 97.28 | 0.9402 | |
Our (MCMS-BCNN-Attention + densenet121) | 98.38 | 99.12 | 97.63 | 98.37 | 0.9560 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 91.36 | 96.20 | 86.13 | 90.89 | 0.8525 |
Dataset | Method | AUC ( Lawn/Mall/City) | AUC Total |
---|---|---|---|
UMN | Wang et al. [41] | 0.9981/0.9724/0.9888 | 0.9864 |
Cong et al. [42] | 0.995/0.975/0.964 | 0.979 | |
Wang et al. [43] | 0.9779/0.9223/0.9849 | 0.9617 | |
Susan et al. [44] | 0.997/0.89/0.962 | 0.945 | |
Zhang et al. [45] | 0.992/0.986/0.979 | 0.985 | |
Kaltsa et al. [46] | 0.9959/0.9338/0.9808 | 0.9702 | |
Xu et al. [47] | 0.9861/0.9864/0.975 | 0.9825 | |
Mu et al. [48] | 0.996/98.3/99.1 | 0.99 | |
Ilyas et al. [49] | 0.9808/0.9955/0.9935 | 0.990 | |
Du Y [50] | 0.998/0.992/0.995 | 0.9950 | |
Sing et al. [51] | 0.9928/0.9885/0.9995 | 0.9936 | |
Our (MCMS-BCNN-Attention + densenet121) | 0.9615/0.9981/0.9056 | 0.9505 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 1.00/99.25/99.33 | 0.9953 | |
PETS2009 | Method | AUC (14-16 View_1) | AUC (14-33 View_1, |
View_2,View_3) | |||
Xu et al. [52] | 0.94 | ||
Zhu et al. (Optical flow) [53] | 0.81 | ||
Zhu et al. (SFM) [53] | 0.94 | ||
Ilyas et al. [49] | - | 0.94 | |
Sing et al. [51] | 0.9788 | - | |
Du Y [50] | 0.9780 | - | |
Our (MCMS-BCNN-Attention + densenet121) | 0.9895 | 0.9666 | |
Our (MCMS-BCNN-Attention+Efficientnetv2) | 1.0 | 1.0 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ekanayake, E.M.C.L.; Lei, Y.; Li, C. Crowd Density Level Estimation and Anomaly Detection Using Multicolumn Multistage Bilinear Convolution Attention Network (MCMS-BCNN-Attention). Appl. Sci. 2023, 13, 248. https://doi.org/10.3390/app13010248
Ekanayake EMCL, Lei Y, Li C. Crowd Density Level Estimation and Anomaly Detection Using Multicolumn Multistage Bilinear Convolution Attention Network (MCMS-BCNN-Attention). Applied Sciences. 2023; 13(1):248. https://doi.org/10.3390/app13010248
Chicago/Turabian StyleEkanayake, E. M. C. L., Yunqi Lei, and Cuihua Li. 2023. "Crowd Density Level Estimation and Anomaly Detection Using Multicolumn Multistage Bilinear Convolution Attention Network (MCMS-BCNN-Attention)" Applied Sciences 13, no. 1: 248. https://doi.org/10.3390/app13010248
APA StyleEkanayake, E. M. C. L., Lei, Y., & Li, C. (2023). Crowd Density Level Estimation and Anomaly Detection Using Multicolumn Multistage Bilinear Convolution Attention Network (MCMS-BCNN-Attention). Applied Sciences, 13(1), 248. https://doi.org/10.3390/app13010248