This project integrates the concepts of Internet of Things with Python. The project was developed thorugh the follwoing phases:
-
A Raspberry Pi was used to capture images of an eomployee who would have access. 30 images were captured using an Open-CV pre-trined classifier. PiRGBArray() gave us a 3-dimensional RGB array from the captures. The advantage is that it has the ability to read from the Raspberry Pi camera as a Numpy array. A directory is created with the name of the employee and the 30 images are stored inside it.
-
A recognizer was trained accoring to the data gathered. the Local Binary Patterns Histogram (LPBH) face recognizer was used. It comes included in the OpenCV package. A dictionary is created which contains the employee name and the Label ID associated with the employee.
-
The recoginzer trained can now be used to detect faces and activate the lock to open the door if the emplyee is recognised by the model. It gives a confidence level for each face it detecs and upon crossing a certain threshold, the lock opens.