Wildfire Detection
Team Members Heading link
- David Conner
- Eric Rivera
- Daood Shah
Project Description Heading link
The challenge we face is that we can’t have a person observing a forest fire due to the large size. We made a video-based wildfire camera detector that has a trained AI. The AI we used was TensorFlow. TensorFlow is a software library for machine learning. It can be used to train deep neural networks, the basis of our design. The design currently consists of a Raspberry Pi (our microcontroller) and a camera that takes in input. We are using a Raspberry Pi because it is a device that is cost-efficient, and it can run our AI, TensorFlow. Our Raspberry Pi is capable of detecting fire, and when flames are present, it informs us that fire is being detected, and when flames are not present, it informs us that flames are not being detected. We did the following test to get a better understanding of what our limitations are for the device—we measured vertical and horizontal distance and blockage tests.