Deprecated: htmlspecialchars(): Passing null to parameter #1 ($string) of type string is deprecated in /home2/muc/public_html/journal/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.php on line 451

Warning: Cannot modify header information - headers already sent by (output started at /home2/muc/public_html/journal/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.php:451) in /home2/muc/public_html/journal/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.php on line 654

Warning: Cannot modify header information - headers already sent by (output started at /home2/muc/public_html/journal/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.php:451) in /home2/muc/public_html/journal/plugins/generic/citationStyleLanguage/CitationStyleLanguagePlugin.php on line 655
TY - JOUR TI - Real-time People Counting with Deep Learning: A Solution for Crowd Management PY - %2025/%08/%25 Y2 - %2025/%12/%22 JF - Al-Mansour Journal JA - مجلة المنصور VL - 42 IS - 1 LA - en KW - Computer vision KW - YOLO KW - Deep learning KW - Python KW - open Cv UR - https://journal.muc.edu.iq/journal/article/view/673 SP - 120-127 AB - Crowd management plays a vital role in ensuring safety, efficiency, and resource optimization in environments such as public events, transportation hubs, shop-ping malls, and workplaces. Traditional people counting methods, including manual observations and sensor-based approaches, often suffer from limitations in accuracy, scalability, and adaptability to dynamic conditions. This research pre-sent a real-time people counting system utilizing deep learning techniques, specifically the YOLO (You Only Look Once) object detection framework. The proposed system processes video streams to accurately detect and count individuals, offering a robust and automated solution for crowd analysis. By leveraging YOLO’s high-speed processing and accuracy, the system effectively identifies people within complex and varying environments, addressing challenges such as occlusions, lighting variations, and high-density crowds. Experimental evaluations demonstrate the system's strong performance, achieving an accuracy of 95% in static environments and 87% in dynamic conditions. These results highlight the model’s reliability, efficiency, and potential applications in real-time crowd monitoring, security surveillance, and urban planning. Future improvements include integrating advanced tracking algorithms and multi-camera setups to enhance detection accuracy and consistency across different perspectives. This research un-derscores the transformative potential of deep learning in crowd management, paving the way for smarter, safer, and more responsive public spaces. ER -