Vision-based Face Recognition using Multi-CNN Algorithm for Hospital Staff Management

Aim: The aim of the study is designed to offer increased security, track attendance, and provide access control within a hospital. Materials and Methods: This project can broadly be categorized into two main groups, namely Group 1 and Group 2. Under Group 1 falls the CNN-based face recognition model that trains on images of hospital staff. Group 2 deals with the Multi CNN model to the real-time hospital management system for identification purposes and tracking the same. A 95% confidence interval has been maintained while keeping the threshold of the performance evaluation at 0.05%. Results: Multi-CNN surpasses the traditional models in face recognition with regards to accuracy and reliability. The model yields a recognition accuracy of 98.7% and also has a comparative decrease in the percentages of false positives and negatives. It has better processing speed and is scalable for the real-time management of the hospital staff. The significance value is 0.0032. Conclusion: Multi-CNN-based face recognition with effectiveness, improved security with smoother flow on administrative procedures above and notably even far greater improvements of all compared with any existing technique within such state-of-the-art in challenging, dynamic environment face recognition environments.