Secure Safes Systems: Combining SSD Face Recognition with One-Time Password Authentication

Secure Safes Systems: Combining SSD Face Recognition with One-Time Password Authentication

연구 분야: Verification

학회: 2024 2nd International Symposium on Information Technology and Digital Innovation (ISITDI)

Safes are specially designed devices for securely storing and protecting documents, valuables, or money. Typically, safes are equipped with locks or digital combinations to prevent unauthorized access. Despite being designed for security, conventional safes are still vulnerable to breaking techniques such as hitting, prying, or using powerful cutting tools. Therefore, this project develops a new system to enhance safe security by implementing dual authentication through face recognition and a one-time password (OTP). This research uses the Single Shot MultiBox Detector (SSD) method for face detection and the Support Vector Machine (SVM) for face classification. The system begins with face recognition of the safe owner, then generates and sends an OTP code input via the keypad. Test results show that the SSD method has 100% accuracy, while the SVM method achieves 87.5% accuracy. The system can generate a six-digit OTP code consisting of numbers and letters, changing every 2 minutes. The SW-420 sensor also successfully detects vibrations with a 100% success rate. All notifications from the system are sent to the user's Telegram application. The results of this study indicate that the proposed approach achieves high accuracy and has the potential to provide a supportive solution for safe security systems.