Main Content

Creating a system in order to detect wildfire using deep learning programmation on a Raspberry Pi 4 and a Wisol module by Sigfox.

Before explaining in details our project and its setup, we wanted to tell you that it is a student project carried out in a shot relatively short time. So we didn’t had enough time to complete our project and to do everything we wanted to. Therefore, our design is incomplete and does not directly answered our problematic. The main purpose of this article is to provide all the resources to anybody who would rebuild our project or continue it and finish it.

Let’s now talk about our project! Like I mentioned before it is a student project for our second year of engineering school in electronics, computer science and embedded systems. Our project’s problematic is:

Design an embedded system, autonomous in energy, user friendly in order to detect wildfires.

To fully answer this issue, we chose to consult an organization expert in wildfire, in order to understand the issues. We’ve contacted “L’Entente Valabre.”

“L’Entente Valabre” is french organism, a clustering of 14 districts (mostly in South of France) which is related to government to help it in protection of forest against fire. They manage formation, prevention, research and development of new way to detect wildfire.

After contacting and discussing with them about needed and restrictions of our problematic. We decided to focus on a system using image processing with deep learning in order to detect wildfire.”

Link to article