Results & Conclusion

Overall, our project was a success! We achieved our original goal of exploring the computational power of the Raspberry Pi and facial recognition classifiers by creating a selective snack dispenser based on whether a person is a friend (known) or foe (unknown). We successfully:

1. Video streamed Pi Camera footage on the PiTFT with boxing of known and unknown faces
2. Dispensed snacks based on whether a person is a friend or foe
3. Took photos on the PiTFT and saved them in a new directory
4. Implemented a large FSM to cycle through different PyGame screens
5. Created complex touchscreen interfaces to receive user input
6. Tangential to the project – we also laser cut for the first time!

From this project, we were able to gain experience working with OpenCV, face_recognition, Raspberry Pi Cameras, servos, and PyGame. This project gave us the freedom to explore these libraries and components in more detail. As a result, we became more familiar with different aspects of the library, while also improving our research skills and ability to read documentation. From this project, we were able to apply a lot of knowledge we had learned throughout the course such as GPIO PWM signals, PyGame, and touchscreen events, and also general practices including modular code structure and iterative testing.