SYSTÈME DE SURVEILLANCE INTELLIGENTE DES FOULES UTILISANT YOLO-GHOST BASÉ SUR L'IOT
DOI :
https://doi.org/10.59277/RRST-EE.2024.69.3.15Mots-clés :
Internet des objets (IoT), Système de surveillance des foules, Apprentissage profond, YOLO-GHOSTRésumé
Les appareils Internet des objets (IoT) offrent une solution de surveillance intelligente, sophistiquée et en temps réel pour les espaces publics. Cependant, en raison des changements instantanés d’éclairage et des angles de vision variables, compter et suivre les personnes dans des scènes bondées constitue un problème difficile. Pour lutter contre ces problèmes, un nouveau YOLO-CROWD est proposé pour un système intelligent de surveillance des foules utilisant YOLO-GHOST. Initialement, une caméra basée sur le protocole Internet (IP) était utilisée pour surveiller et capturer les foules dans une séquence vidéo. Les séquences vidéo capturées sont converties en images et transmises au serveur via Internet. Les images enregistrées sont transmises au module de classification YOLO-GHOST pour effectuer la détection et le comptage des personnes. Enfin, les résultats de sortie détectés sont transférés au serveur du centre de surveillance. La technique YOLO-CROWD est simulée à l'aide de MATLAB. L'efficacité de la technique YOLO-CROWD proposée est évaluée à l'aide de mesures d'évaluation telles que l'exactitude, la précision, le rappel, la sensibilité, le score F1 et la précision moyenne moyenne. Les résultats expérimentaux montrent que la précision du YOLO-CROWD a augmenté jusqu'à 99,95 %, prouvant que son utilisation prévue est une détection précise des foules. La précision de détection de la méthode proposée est respectivement de 84,9 %, 87,58 %, 93,91 % et 97,72 % supérieure à celle des systèmes EABeD, LCDnet, CDEM-M et Public Vision existants.
Références
(1) N. Thakur, P. Nagrath, R. Jain, D. Saini, N. Sharma, D.J. Hemanth, Autonomous pedestrian detection for crowd surveillance using deep learning framework, Soft Computing, 27, 14, pp. 9383-9399 (2023).
(2) M.Ş. Gündüz, G. Işık, A new YOLO-based method for real-time crowd detection from video and performance analysis of YOLO models. Journal of Real-Time Image Processing, 20, 1 (2023).
(3) A.K. Suhane, A.V.H.P.U. Raghuwanshi, A. Nimbark, L. Saxena, Human detection and crowd counting using Yolo, Conference: Tech Fest IIT Roorkee (2023).
(4) S. Zheng, J. Wu, S. Duan, F. Liu, J. Pan, An improved crowd counting method based on YOLOv3, Mobile Networks, and Applications, pp. 1–9 (2022).
(5) S. Mohanapriya, P. Natesan, K. Rinisha, S. Nishanth, J. Robin, Video segmentation using YOLOv5 for surveillance, 2nd International Conference on Vision Towards Emerging Trends in Communication and Networking Technologies (ViTECoN), pp. 1–5 (2023).
(6) F. Liu, Z. Han, H. Song, J. Wang, C. Liu, G. Ban, Crowd sensing and spatiotemporal analysis in urban open space using multi‐viewpoint geotagged videos, Transactions in GIS, 27, 2, pp. 494-515 (2023).
(7) A. Alamri, Cloud of things in crowd engineering: a tile-map-based method for intelligent monitoring of outdoor crowd density, Sensors, 22, 9, pp. 3328 (2022).
(8) S. Wang, Z. Pu, Q. Li, Y. Wang, Estimating crowd density with edge intelligence based on lightweight convolutional neural networks, Expert Systems with Applications, 206, pp. 117823 (2022).
(9) T. Motedayen, M. Yaghoobi, M. Kheirabadi, Improving mobile mass monitoring in the IoT environment based on Fog computing using an improved forest optimization algorithm, Journal of Control and Decision, pp. 1–14 (2022).
(10) M.P. Kantipudi, R. Aluvalu, S. Velamuri, An intelligent approach of intrusion detection in mobile crowd sourcing systems in the context of IoT based SMART city, Smart Science, 11, 1, pp. 234–240 (2023).
(11) F. Aldosari, S. Alamri, R. Alharbi, G. Alghamdi, IoT based crowd management tool, Proceedings of SAI Intelligent Systems Conference, pp. 533–539 (2022).
(12) A.I. Middya, P. Dey, S. Roy, IoT-based crowdsensing for smart environments, Internet of Things for Smart Environments, pp. 33–58), Springer International Publishing (2022).
(13) H. Zhao, W. Min, J. Xu, Q. Wang, Y. Zou, Q. Fu, Scene-adaptive crowd counting method based on meta learning with dual-input network DMNet. Frontiers of Computer Science, 17, 1, 171304 (2023).
(14) L. Zhang, L. Yan, M. Zhang, J. Lu, T2CNN: a novel method for crowd counting via two-task convolutional neural network, The Visual Computer, 39, 1, pp. 73–85 (2023).
(15) S.S. Savner, V. Kanhangad, CrowdFormer: weakly-supervised crowd counting with improved generalizability, Journal of Visual Communication and Image Representation, 94, 103853 (2023).
(16) I. Ahmed, M. Ahmad, A. Ahmad, G. Jeon, IoT-based crowd monitoring system: using SSD with transfer learning, Computers & Electrical Engineering, 93,107226 (2021).
(17) W. Xiao, M. Li, B. Alzahrani, R. Alotaibi, A. Barnawi, Q. Ai, A blockchain-based secure crowd monitoring system using UAV swarm. IEEE Network, 35, 1, pp. 108–115 (2021).
(18) L. Rajendran, R.S. Shankaran, Bigdata enabled realtime crowd surveillance using artificial intelligence and deep learning, IEEE International Conference on Big Data and Smart Computing (BigComp), pp. 129–132 (2021).
(19) M.U. Farooq, M.N.M. Saad, S.D. Khan, Motion-shape-based deep learning approach for divergence behavior detection in high-density crowd, The Visual Computer, 38, 5, pp.1553–1577 (2022).
(20) H. Choi, M. Fujimoto, T. Matsui, S. Misaki, K. Yasumoto, Wi-cal: Wifi sensing and machine learning based device-free crowd counting and localization, IEEE Access, 10, pp. 24395–24410 (2022).
(21) S. Wang, Z. Pu, Q. Li, Y. Wang, Estimating crowd density with edge intelligence based on lightweight convolutional neural networks, Expert Systems with Applications, 206, 117823 (2022).
(22) Y. Zhen, M. Sugasaki, Y. Kawahara, K. Tsubouchi, M. Ishige, H. Murakami, M. Shimosaka, Efficient adaptive beacon deployment optimization for indoor crowd monitoring applications, Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 6, 4, pp. 1–22 (2023).
(23) M.A. Khan, H. Menouar, and R. Hamila, LCDnet: a lightweight crowd density estimation model for real-time video surveillance, Journal of Real-Time Image Processing, 20, 2, 29 (2023).
(24) X. Zhang, Y. Sun, Q. Li, X. Li and X. Shi, Crowd density estimation and mapping method based on surveillance video and GIS, ISPRS International Journal of Geo-Information, 12, 2, 56 (2023).
(25) M. Qaraqe, A. Elzein, E. Basaran, Y. Yang, E.B. Varghese, W. Costandi, J. Rizk, N. Alam, PublicVision: a secure smart surveillance system for crowd behavior recognition, IEEE Access, 12, pp. 26474-26491 (2024).
(26) G.C. Serițan, B.A. Enache, I. Vilciu, S.D. Grigorescu, V. Mladenov, Comparison study of top development boards in the context of Iot, Rev. Roum. Sci. Techn. – Électrotechn. Et Énerg., 67, 4, pp. 483–486 (2022).
(27) R.R. Sathiya, S. Rajakumar, J. Sathiamoorthy, Secure blockchain-based deep learning approach for data transmission in IOT-enabled healthcare system, International Journal of Computer and Engineering Optimization, 01, 01, pp. 15–23 (2023).
(28) S. Ramamoorthi, A. Appathurai, Energy-aware clustered blockchain data for IoT: an end-to-end lightweight secure & Enroute filtering approach, Computer Communications, 202, pp. 166 –182 (2023).
(29) S. Reeba Rex, T. Pravin Rose, S. Amudaria, Real-time remote monitoring via horse head optimization deep learning network, International Journal of Data Science and Artificial Intelligence, 02, 02, pp. 42–47 (2024).
(30) R. Sundarasekar, A. Appathurai, Automatic brain tumor detection and classification based on IoT and machine learning techniques, Fluctuation and Noise Letters, 21, 03, 2250030 (2022).
(31) R.A. Crăciun, R.N. Pietraru, M.A. Moisescu, Internet of Things Platform Benchmark: An Artificial Intelligence Assessment, Rev. Roum. Sci. Techn. – Électrotechn. Et Énerg., 69, 1, pp. 97–102 (2024).
Téléchargements
Publiée
Numéro
Rubrique
Licence
(c) Copyright REVUE ROUMAINE DES SCIENCES TECHNIQUES — SÉRIE ÉLECTROTECHNIQUE ET ÉNERGÉTIQUE 2024
Ce travail est disponible sous licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.
