An Improved way of the token-based system in hostel using advanced technique of computer vision which helps us to train people image/face in the real-time and to have training internally with the specific id so that we can take attendance in real-time and python script will automatically print the tokens based on the update of recognized attendance in the comma-separated database file. This system takes one level advance in making only the recognized people get the token without the intervention of security or a separate staff to monitor the people. We have used Python 3.8 version and the main technology behind the scene is the computer vision technology which enables us to understand the pattern of a human face structure so that we can train and use it for future recognition activities. With the advancement of technology, we came up with an automatic token vending system based on recognized people of the community or organization. We also have built a GUI TKinter application for better usability.

Keywords : Computer Vision, Python, Training, Face Haar Cascade, GUI, Tkinter.