AutoLogger

Autonomous attendance logger designed and developed for the HSRW Robotics lab.

The Raspberry Pi uses face recognition to detect faces and recognise from existing database. Then, uses hand gesture detection to log in/out.

This fork presents a GUI that should make it easier to use the autologger with the Touchscreen.

Hardware

• Raspberry Pi 3
• Raspberry Pi Camera V2
• Raspberry Pi 7" Touchscreen with enclosure

Data Logging

• Text files are created daily to store check-in and check out of users with timestamps. The logging directory maintains a hierarchy based on year and month.
• The logbook folder must be created manually, otherwise the code returns an error.
• System requires a virtual keyboard installed for new username input.

sudo apt-get install matchbox-keyboard

GUI

• A simple standby menu created using pygame and pygameMenu.

pip install pygame-menu

• Responds to mouse, keyboard and touch screen.

TODO

• Automate new user creation process
• Make system "hands-free"; may require more hardware
• Avoid writing images to file: ?
• Need to add code so that database is created if it doesn't exist
• Pressing the buttons multiple times sets up the event every time; which needs to be sorted out