Food Device Challenge: h0t_d0g

h0t_d0g is a state-of-the-art device that solves a pressing issue in our society: hot dog detection. Using a Raspberry Pi 3B+, a Raspberry Pi HQ camera module, and a pretrained convolutional neural network called YOLOv8, we have been able to create a device that can tell you if an object is:

  1. A hog dog
  2. NOT a hot dog

The design also features a custom Raspberry Pi hat that includes a button for triggering the camera module, and an OLED screen, LEDs, buzzers, and vibrational motors for indication.

Yes, we are aware this device is similar to a bit from Silicon Valley. Consider this another beautiful example of simultaneous invention.

The setup for the scripts is unfortunately not very straightforward, so we recommend just using a copy of our Raspberry Pi image if you would like to demo the device yourself.

A demo of our device can be seen on YouTube here.

The GitHub repo contains design files for the Raspberry Pi hat, copies of the shell and python scripts we are running and a high level BOM.

1 Like

Thanks for the entry @ggiannoulis ! Love the custom PCB Shield!