Deep Learning Based Efficient Crowd Counting System
Source abstract
Abstract
Estimation of crowd count is becoming crucial nowadays, as it can help in security surveillance, crowd monitoring, and management for different events. It is challenging to determine the approximate crowd size from an image of the crowd’s density. Therefore in this research study, we proposed a multi-headed convolutional neural network architecture-based model for crowd counting, where we divided our proposed model into two main components: (i) the convolutional neural network, which extracts the feature across the whole image that is given to it as an input, and (ii) the multi-headed layers, which make it easier to evaluate density maps to estimate the number of people in the input image and determine their number in the crowd. We employed the available public benchmark crowd-counting datasets UCF CC 50 and ShanghaiTech parts A and B for model training and testing to validate the model’s performance. To analyze the results, we used two metrics Mean Absolute Error (MAE) and Mean Square Error (MSE), and compared the results of the proposed systems with the state-of-art models of crowd counting. The results show the superiority of the proposed system.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Video Surveillance and Tracking Methods · Anomaly Detection Techniques and Applications · Evacuation and Crowd Dynamics
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (54)
Show source-provided references
1. Anomalous entities detection and localization in pedestrian flows10.1016/j.neucom.2018.02.045 ↗
2. A smartphone-based adaptive recognition and real-time monitoring system for human activities10.1109/THMS.2020.2984181 ↗
3. Deep neural network approach in robot tool dynamics identification for bilateral teleoperation10.1109/LRA.2020.2974445 ↗
4. Detecting pedestrians by learning shapelet features
5. Detection of multiple, partially occluded humans in a single image by bayesian combination of edgelet part detectors
6. Histograms of oriented gradients for human detection
7. Object detection with discriminatively trained part-based models10.1109/TPAMI.2009.167 ↗
8. Object detection in high-resolution remote sensing images using rotation invariant parts based model10.1109/LGRS.2013.2246538 ↗
9. Bayesian poisson regression for crowd counting
10. Cost-sensitive sparse linear regression for crowd counting with imbalanced training data
11. An evaluation of crowd counting methods, features and regression models10.1016/j.cviu.2014.07.008 ↗
12. Multi-source multi-scale counting in extremely dense crowd images
13. Learning to count objects in images
14. Single-image crowd counting via multi-column convolutional neural network
15. CrowdNet: A deep convolutional network for dense crowd counting
16. Very deep convolutional networks for large-scale image recognition
17. Towards perspective-free object counting with deep learning
18. Switching convolutional neural network for crowd counting
19. CNN-based cascaded multi-task learning of high-level prior and density estimation for crowd counting
20. Attention is all you need
21. Training tips for the transformer model10.2478/pralin-2018-0002 ↗
22. F. Chollet, “Keras: The python deep learning library, github repository,” 2015. Accessed: Dec. 15, 2023. [Online]. Available: https://keras.io/
23. ImageNet classification with deep convolutional neural networks,” presented at the 25th
24. EfficientNet: Rethinking model scaling for convolutional neural networks
25. Crowd monitoring using image processing10.1049/ecej:19950106 ↗
26. Fast neural learning vision system for crowd estimation at underground stations platform10.1023/A:1018781301409 ↗
27. Estimation of crowd density based on wavelet and support vector machine10.1191/0142331206tim178oa ↗
28. Counting people using video cameras10.1080/17445760601139096 ↗
29. Privacy preserving crowd monitoring: Counting people without people models or tracking
30. Analysis of crowded scenes using holistic properties
31. A method for counting people in crowded scenes
32. A method for people counting using feature fusion based on SVR with PSO optimization
33. Feature mining for localised crowd counting
34. End-to-end crowd counting via joint learning local and global count
35. Generating high-quality crowd density maps using contextual pyramid CNNs
36. MH-MetroNet–A multi-head CNN for passenger-crowd attendance estimation10.3390/jimaging6070062 ↗
37. Relational attention network for crowd counting
38. Crowd detection and analysis for surveillance videos using deep learning
39. Crowd density estimation using deep learning for Hajj pilgrimage video analytics10.12688/f1000research.73156.1 ↗
40. Generative adversarial network based abnormal behavior detection in massive crowd videos: A Hajj case study10.1007/s12652-021-03323-5 ↗
41. CSRNet: Dilated convolutional neural networks for understanding the highly congested scenes
42. Densely connected convolutional networks
43. Squeeze-and-excitation networks10.1109/TPAMI.2019.2913372 ↗
44. Deep residual learning for image recognition
45. Scale aggregation network for accurate and efficient crowd counting
46. Crowd counting via scale-adaptive convolutional neural network
47. Crowd counting via adversarial cross-scale consistency pursuit
48. PaDNet: Pan-density crowd counting10.1109/TIP.2019.2952083 ↗
49. Pushing the frontiers of unconstrained crowd counting: New dataset and benchmark method
50. Adaptive density map generation for crowd counting
51. Estimation of the number of passengers in a bus using deep learning10.3390/s20082178 ↗
52. Sensing technologies for crowd management, adaptation, and information dissemination in public transportation systems: A review10.1109/JSEN.2022.3223297 ↗
53. An efficient management platform for developing smart cities: Solution for real-time and future crowd detection10.3390/electronics10070765 ↗
54. Routing for crowd management in smart cities: A deep reinforcement learning perspective10.1109/MCOM.2019.1800603 ↗