The objective of this research project is to develop a PDS (Pedestrian Detection System) to quickly and reliably identify the presence of a pedestrian from one or more cameras.

This PDS must be functional during the night thanks to artificial intelligence algorithms making it possible to classify objects with a parallel implementation.

The overall goal is to reduce the impact of urban lighting on biodiversity and save energy by detecting human activities to turn public lighting on or off when it seems relevant. The images are retrieved using a system consisting of a Raspberry microcontroller and a Raspberry Pi infrared camera which are processed on the fly to detect the presence of a pedestrian. The lighting can then be dynamically controlled thanks to this result.