The pervasive use of mobile phones in designated no-mobile zones, including hospitals, gas stations, libraries, and pedestrian crossings, poses significant safety risks and regulatory challenges. To address these issues, this paper introduces an advanced detection system leveraging the YOLOv8x deep learning architecture. This system is designed to automatically identify individuals using mobile phones in restricted areas, utilizing CCTV and surveillance footage for data collection. The images undergo extensive preprocessing, including resizing, normalization, and noise reduction, to prepare them for effective model training. The YOLO8x model, known for its efficiency in real-time object detection, is trained to accurately detect mobile phone users by learning from annotated datasets. Evaluation metrics, such as Precision, Recall, and mean Average Precision (mAP), are used to evaluate the system’s accuracy in identifying violations. High precision indicates the system’s ability to minimize false positives, while high recall reflects its capacity to detect most violations. The system’s real-time capabilities allow for immediate alerts and interventions, enhancing public safety and regulatory compliance. The results demonstrate the system’s robustness and reliability, offering significant potential for deployment in various sensitive environments to reduce disruptions and ensure a safer public space. This detection solution enforces no-mobile zone policies and creates a safer environment, thereby reducing the risks of unauthorized mobile phone use.