Intelligence Surveillance System for Bank Security Against Robbery

The role of banks in the economic growth of a country is very crucial and there is always a concern about the security of money and valuable items kept. Bank robbery is often accompanied by the use of force, violence, and threats. To address this issue, the proposed work aims to design an Intelligence Surveillance System for Bank Security against robbery. The system automatically detects suspicious activities, such as the presence of weapons, using the YOLO architecture of Convolutional Neural Networks (CNNs). The methodology involves training the model on a comprehensive dataset of suspicious activities in banks and fine-tuning the model using transfer learning techniques. Once the system detects such suspicious activity, the intelligent surveillance system promptly triggers an automated alert system. The system notifies the nearby police station by generating an immediate SMS, alerting them to the ongoing robbery. Furthermore, the system captures and transmits relevant images of the robbery to provide visual evidence to law enforcement. The performance of the intelligent surveillance system is evaluated based on key metrics such as accuracy, precision, recall, and F1-score. The algorithm is tested on different training testing ratios such as 90–10, 80–20, 70–30, and 60–40. The highest accuracy of 91%, precision of 95%, recall of 98%, and F1-score of 95% are achieved with an 80:20 ratio. The resulting system aims to provide an effective and reliable solution for enhancing bank safety and reducing the impact of bank robberies on both financial institutions and public safety.