Real-Time Mobile Detection to Prevent Cheating in Examinations

The widespread use of mobile phones in examination settings has become a significant concern, facilitating cheating and compromising the integrity of assessments. Mobile devices, with their access to online resources, communication platforms, and stealthy applications, pose a serious threat to maintaining a fair evaluation process. This paper presents a deep learning-based model designed for the real-time detection of mobile devices during examinations. The model is capable of accurately identifying mobile devices, including Android smartphones, tablets, and feature phones, while distinguishing them from other common items found in examination settings, such as calculators, books, notebooks, pocket diaries, and sunglasses. The proposed model is trained on a curated dataset, combining the Kaggle Driver dataset with a custom-built Classroom Resource Dataset (CRD) that includes labelled data of students in typical examination scenarios, both with mobile devices and permissible items. The model exhibits robust object classification capabilities, providing a reliable tool to enhance examination security by preventing the use of unauthorized mobile devices.