Predicting butterfly species presence from satellite imagery using soft contrastive regularisation

2025-05-14

AI Summary

This study addresses the challenge of presence-only prediction for multiple co-occurring butterfly species in satellite imagery, motivated by large-scale biodiversity monitoring requirements. We introduce the first remote-sensing dataset annotated for butterfly species across the UK, featuring four spectral bands, and develop a ResNet-based multi-label classification framework. To enhance discrimination among co-occurring species—particularly under probabilistic labels (e.g., species occurrence probabilities)—we propose a novel soft-supervised contrastive regularization loss. Experimental results demonstrate that our method significantly outperforms mean-baseline approaches in high-biodiversity regions, achieving improved accuracy for species coexistence modeling. This work validates the feasibility and effectiveness of remote-sensing-driven species distribution modeling and establishes a scalable technical pathway for ecological monitoring.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

The growing demand for scalable biodiversity monitoring methods has fuelled interest in remote sensing data, due to its widespread availability and extensive coverage. Traditionally, the application of remote sensing to biodiversity research has focused on mapping and monitoring habitats, but with increasing availability of large-scale citizen-science wildlife observation data, recent methods have started to explore predicting multi-species presence directly from satellite images. This paper presents a new data set for predicting butterfly species presence from satellite data in the United Kingdom. We experimentally optimise a Resnet-based model to predict multi-species presence from 4-band satellite images, and find that this model especially outperforms the mean rate baseline for locations with high species biodiversity. To improve performance, we develop a soft, supervised contrastive regularisation loss that is tailored to probabilistic labels (such as species-presence data), and demonstrate that this improves prediction accuracy. In summary, our new data set and contrastive regularisation method contribute to the open challenge of accurately predicting species biodiversity from remote sensing data, which is key for efficient biodiversity monitoring.

Problem

• Predict butterfly species presence from satellite imagery
• Optimize Resnet model for multi-species presence prediction
• Develop soft contrastive regularization for probabilistic labels

Innovation

• Resnet-based model for multi-species presence prediction
• Soft supervised contrastive regularization loss
• 4-band satellite imagery for biodiversity monitoring