Transformative potential of digital systems for promoting human-wildlife coexistence: A systematic literature review

Bas S.H.T. Michielsen Gerard Schouten Joris P. G. Cromsigt Remco C. Veltkamp Irma Arts Jitse Bijlmakers Magali Frauendorf Tim R. Hofmeester Amy Newsom Minh-Xuân A. Truong Helen Verploegen Els Weinans Wenjin Yu Silja Zimmermann Ine Dorresteijn

Digital technology plays an increasingly important role in wildlife management and conservation, by enhancing monitoring capabilities and reshaping human-wildlife interactions. However, the transformative potential of these digital solutions for coexistence remains unclear. This paper presents a novel framework to assess the transformative potential of digital systems in wildlife management and conservation, focusing on two key factors: Digital Maturity, which evaluates technical sophistication of digital systems, and Systemic Depth, which measures their capacity for enabling lasting change. We used this framework to evaluate 524 studies in a systematic literature review in wildlife management and conservation, and found that although sometimes higher Digital Maturity or Systemic Depth was achieved, overall the transformative potential of applied digital systems was still low. Studies that scored high emphasize interdisciplinary collaboration, adaptability, data sharing, and technologies such as machine learning. This research highlights achieving transformative potential requires a holistic approach integrating ecological, social, and technological perspectives.

DOI Dataset Code WildlifeNL

Data-centric AI approach for automated wildflower monitoring

Gerard Schouten Bas S.H.T. Michielsen Barbara Gravendeel

We present the Eindhoven Wildflower Dataset (EWD) as well as a PyTorch object detection model that is able to classify and count wildflowers. EWD, collected over two entire flowering seasons and expert annotated, contains 2,002 top-view images of flowering plants captured ‘in the wild’ in five different landscape types (roadsides, urban green spaces, cropland, weed-rich grassland, marshland). It holds a total of 65,571 annotations for 160 species belonging to 31 different families of flowering plants and serves as a reference dataset for automating wildflower monitoring and object detection in general. To ensure consistent annotations, we define species-specific floral count units and provide extensive annotation guidelines. With a 0.82 mAP (@IoU > 0.50) score the presented baseline model, trained on a balanced subset of EWD, is to the best of our knowledge superior in its class. Our approach empowers automated quantification of wildflower richness and abundance, which helps understanding and assessing natural capital, and encourages the development of standards for AI-based wildflower monitoring. The annotated EWD dataset and the code to train and run the baseline model are publicly available.

DOI Dataset Code FlowerPower

Education

PhD

2023 to present, Sustainable Development

Utrecht University, Utrecht, Netherlands

MSc Information Systems

2006 to 2008, School of Business and Social Sciences

University of Roehampton, London, United Kingdom

BSc Mens & Informatica

2002 to 2006, School of ICT

Fontys University of Applied Sciences, Eindhoven, Netherlands

Experience