Deep learning-based hierarchical insect classification using camera trap imagery

2026-07-30

AI Summary

This study addresses the challenges in monitoring insect biodiversity—namely, the scarcity of standardized data, high costs of manual identification, and the highly imbalanced nature of hierarchical classification—by proposing a deep learning–driven hierarchical image classification system. Leveraging a large-scale, long-tailed dataset of non-lethally captured insect images from camera traps, the system performs automated identification across a five-level taxonomy comprising 34 classes. The method introduces a novel architecture adaptable to variable taxonomic depths, integrating class-balanced weighting with a confidence-threshold fallback mechanism to ensure taxonomically consistent and granularity-adaptive predictions. Experimental results demonstrate per-level accuracies ranging from 80% to 99%, substantially outperforming non-hierarchical baselines and confirming the approach’s efficacy and robustness in real-world scenarios.

Abstract

Declining insect populations make reliable biodiversity monitoring increasingly urgent, yet monitoring of insect biodiversity is hampered by a lack of standardised data and by costly and time-consuming manual identification by expert entomologists. Deep learning-based image classifiers, processing data from automated non-lethal camera traps, have the potential to transform and scale insect biodiversity monitoring. However, challenges remain in acquiring expert-annotated datasets, developing model architectures that generalise well across diverse taxonomic levels and training models on highly imbalanced data. Hierarchical data also benefits from designing models that default to higher-confidence, coarser-level predictions, when uncertain about finer taxonomic levels. In this paper we address these challenges with a deep learning-based hierarchical classification model. First, we present a manually curated, long-tailed dataset of around one million images of insects, extracted from 1,801 camera-trap video recordings and annotated with a five-level, 34-class hierarchy. Further, we adapt a hierarchical classification model architecture to a five-level variable-depth hierarchy, with class-balanced weighting. Our model improves on non-hierarchical classifiers by leveraging biological taxonomy to extract granularity-specific visual features and makes hierarchy-consistent predictions to the deepest taxonomic level that meets a confidence threshold (T = 0.6). Our model achieved a per-level accuracy of 80-99% on test data, across five levels of hierarchy. Furthermore ...

Problem

• insect biodiversity monitoring
• camera trap imagery
• hierarchical classification
• imbalanced data
• expert annotation

Innovation

• hierarchical classification
• camera trap imagery
• long-tailed dataset
• taxonomy-aware deep learning
• confidence-thresholded prediction