Towards an Automated Biodiversity Modelling Process for Forest Animals using Uncrewed Aerial Vehicles

Climate change poses a grave threat to habitats such as forests, endangering the integrity and biodiversity of the global flora and fauna. Accurate surveying techniques are crucial to model populations, detect over and under populations, and address them accordingly. This work proposes a process for creating a biodiversity model of a forest’s fauna using uncrewed aerial vehicles equipped with RGB and thermal cameras. Real-world data, combined with computer-generated imagery and artificial intelligence models, will allow training suitable computer vision models. These models will serve as a reliable and objective data source, enabling the creation of statistical models to describe the monitored forests’ conditions and the biodiversity of its fauna. The proposed methodology is expected to have significant implications for conservation efforts. It should represent a reliable and efficient way to monitor and evaluate forest ecosystems, identifying areas of concern and prioritizing conservation efforts. By providing a comprehensive understanding of the biodiversity within a forest, it could help policymakers make informed decisions and develop effective conservation strategies. Ultimately, this work aims to contribute to the preservation of our planet’s biodiversity and the protection of its habitats.