Case Study: Facial Recognition for Building Entry
Redefining Access in post covid world
Background A large corporate company needed a way to monitor the people entering their building through the main entrance using CCTV footage.
Problem Statement The challenge was to identify and classify the faces of people entering the building as either known or unknown.
Solution Approach Our team developed a solution using an open source framework for facial recognition. We implemented a deep learning algorithm based on a convolutional neural network (CNN) to recognize faces.
The solution captured frames from the CCTV footage, extracted faces from each frame, and then compared the faces against a database of known faces pre-registered in the system. If a match was found, the person was classified as "known." If no match was found, the person was classified as "unknown."
Our team used a face recognition API to integrate the system with the existing CCTV infrastructure, and the data of recognized people was stored in a database for up to one month for analysis.
Benefits The solution provided several benefits for the client. Firstly, it improved security and access control by allowing the client to track and monitor the people entering the building. Secondly, it enabled visitor management by keeping a record of who entered the building and at what time. Lastly, it could be used for attendance tracking purposes for employees.
Conclusion Our facial recognition solution provided an accurate and efficient way to classify people entering the building into known and unknown categories. It improved security, visitor management, and employee attendance tracking. The solution was easy to integrate with the existing infrastructure and could be easily analyzed for further insights.