AI-Powered Plant Species Mapping to Support Phenology and Climate Research

Phenology, the study of recurrent life cycles events and its relationship to climate, is a key discipline in climate change research. A fundamental requirement for phenology studies is the identification of plant species in images captured by unmanned aerial vehicles and tower-based near-surface technologies as they provide key indicators of ecosystem health and biodiversity. Last years, thanks to its great learning capacity from data exposure, deep learning has led to remarkable progress in the field of computer vision. Largely, this is due to the availability of large amounts of labeled data that has contributed to the development of models with extraordinary inference capabilities. Acquiring and annotating a desirable amount of data to satisfy these models is often a hard task, requiring overwhelming human effort and specific expertise. This reliance on exhaustive labeling is a key limitation to deploying systems capable of identifying plant species in the real world. Although deep learning models excel in many tasks, occasionally beating humans, their generalization ability is poor, only performing tasks they are trained for. The ability to adapt to novel scenarios is the hallmark of human intelligence. Motivated by such aspects, this research proposal aims to investigate methods to improve the generalization ability of deep learning, enabling to identify plant species with minimal human supervision. For this, we intend to advance the state of the art in model generalization, aiming to learn with most informative data and put the human in the loop in a more effective manner. Finally, the results of this research proposal are also meant to contribute to the understanding of the impact of climate change on vegetation phenology.