Prototype Attendance Module for Tutoring Sessions
The clientside code contains a UI for users to signin and out with. It uses OpenCV to poll for faces through a webcam. It then sends the images to the server to attempt recognize a user that used the service in the past and set up a data set (14 images) for recognition. It also sends a signal to server when closed so that the server may update a spreadsheet (kept on google sheets) to keep track of user data.
The four functions of the ui are:
- Sign users in
- Sign users out
- Poll for facial recognition
- Update Spread sheet upon closing
Users are presented with a basic sign in screen on top left of module. There they will enter there first and last name and course in attendance for. Then they can just click the signin/out button. They will be signed in or out and the time recorded for the event on the server side as most attendance works.
Using OpenCV I wanted to make that process even faster or at least more automated. Therefore, a user can create a recognition data set by clicking create dataset and 14 images will be captured of there face. Of course the user will verify each one to help ensure a proper dataset was made. Then, they will be signed in upon a successful creation of the data set. The video recording may be seen in the top right while confirmation images are seen in the bottom right.
Instructions for use are found in the bottom left. Confirmation text of any event is found in the box where the action happens.