2nd European GREEN Conference – EGC 2024
11-14 June 2024 | Vodice, CROATIA
Plenary Speaker
Prof. Dr. IVANA MAJIĆ
University of Osijek, Faculty of Agrobiotechnical Sciences Osijek (Croatia)
Ivana Majić is a Full Professor in Biotechnical Sciences at the Faculty of Agrobiotechnical Sciences Osijek, Josip Juraj Strossmayer University of Osijek, Croatia. She holds a PhD in Plant Nematology and. As Vice-dean for International Cooperation and Studies in English, she oversees global partnerships and teaches courses on Entomology, Nematology, and Integrated Plant Protection. Her research focuses on environmental protection, biological and chemical pest control, host-parasite relationships, and multitrophic interactions among invertebrates. She has contributed to several international research projects and has co-authored many publications. She established a multidisciplinary team with the aim of developing innovative methods using digital technologies to conserve biodiversity, demonstrating her commitment to sustainability.
Plenary lecture:
AI orchestrates biodiversity monitoring in Nature Park Kopački Rit
Monitoring biodiversity is crucial for understanding global changes caused by human activity and climate change. Technological advancements, open science, and collaborative efforts in the past decade promise global access to rapid and affordable biodiversity monitoring tools.
Researchers have long used sound to study wildlife, now often employing audio recording technology instead of relying solely on identifying species by ear. Sound is crucial for vocal animals in communication, mating, navigation, and territorial defense. Bioacoustics and ecoacoustics are used to monitor biodiversity of the animal community, including insect, bird, amphibian, mammal, fish, and bat species. Passive acoustic monitoring devices are easy to deploy and can operate for long durations, providing valuable insights into habitats, animal behaviors, and potential illegal activities. Although this technology offers considerable advantages, researchers find the processing of the substantial amounts of generated data time-consuming. Deep learning algorithms have greatly advanced bioacoustics research, with convolutional neural networks being prominently featured in recent scientific articles on bioacoustic classification models.
In this talk, we outline the bioacoustic monitoring project’s in Nature Park Kopački Rit and its current developments, also an approach that utilizes passive acoustic monitoring and machine learning techniques to automatically extract features from time-series audio signals and employ deep learning models for classifying different animal species based on their vocalizations.
The project has been granted under the Tech4All program.
Keywords: Wildlife identification; Bioacoustics; Passive acoustic monitoring; Machine learning
BACK
IAESP 