The Sonoma State Smart Pig Trap is a project by Edwin Tran, Richard Mitts and Taylor Plorin. It is designed as an end to end system that will detect motion inside of an animal cage, and verify that there is something in the cage, before setting off a camera. A picture of the animal in the cage is ran through the Tensorflow Inception v3 library, and if it matches with the desired object, the solenoid will activate, and the cage will close. Otherwise it will remain open. Given a wifi connection, the trap will also inform users via Twitter and email about the animals that have been caught in the cage, as well as send periodic updates.
The device is designed to run for long periods of time outdoors with the help of a battery and a 9W solar panel, all enclosed in a weather-proof case, and is designed to be adaptable to different cage and gate types to suit your purpose.
Copyright © 2019 Sonoma State Smart Animal Trap - All Rights Reserved.