Academic Journal of Engineering and Technology Science, 2024, 7(6); doi: 10.25236/AJETS.2024.070609.
Zhang Jiang, Lv Kunshan, Tan Yao, Yu Jingtong
School of Electrical Engineering, Northwest Minzu University, Lanzhou, 730124, China
In response to the severe situation of frequent accidents in crowded places globally in recent years, this study aims to develop an advanced intelligent crowd evacuation system. This paper employs an improved Multi-Column Convolutional Neural Network (MCNN) algorithm for deep learning to segment crowds from images for subsequent feature extraction. Through data sampling, training, and analysis, the images are mapped to crowd density maps, enabling the system to prevent excessive crowd density in advance. Additionally, in scenarios of overcrowding, where there is a lack of real-time planning for escape routes in densely populated areas and a low success rate of evacuation, a dynamic and real-time escape route indication system is designed. This system utilizes monitoring data and the Dijkstra dynamic optimization algorithm to calculate the shortest path, directing crowd evacuation based on the principles of optimality and speed. Simulation data demonstrates that this can improve the effectiveness of crowd evacuation.
Intelligent Crowd Evacuation System, Image Recognition, Deep Learning Algorithm, Crowd Density Monitoring, Evacuation Route Planning
Zhang Jiang, Lv Kunshan, Tan Yao, Yu Jingtong. Modeling and Verification of Intelligent Crowd Evacuation Systems. Academic Journal of Engineering and Technology Science (2024) Vol. 7, Issue 6: 62-68. https://doi.org/10.25236/AJETS.2024.070609.
[1] Zhang, Y., Zhou, D., Chen, S., & Bai, X. (2015). Single-Image Crowd Counting via Multi-Column Convolutional Neural Network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1307-1315).
[2] Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). SSD: Single Shot MultiBox Detector. In European Conference on Computer Vision (pp. 21-37). Springer, Cham.
[3] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks. In Advances in Neural Information Processing Systems (pp. 1097-1105).
[4] Helbing, D., Farkas, I., & Vicsek, T. (2000). Simulating Dynamical Features of Escape Panic. Nature, 407(6803), 487-490.
[5] Löhner, R., & Kress, O. (2003). Finite Element Simulation of Pedestrian Flows. Pedestrian and Evacuation Dynamics, 2003, 1-14.