Overview of LifeCLEF Plant Identification task 2019: diving into data deficient tropical countries

2025-09-23

AI Summary

Tropical regions—such as the Guiana Shield and northern Amazon rainforest—suffer from severe scarcity of plant occurrence data, and existing automated identification models exhibit poor generalization across ecological and geographic contexts. Method: LifeCLEF 2019 Plant Identification Challenge addressed this by constructing a large-scale image dataset covering ~10,000 tropical plant species and establishing the first cross-regional, multi-species, fine-grained recognition benchmark under realistic ecological conditions. It systematically evaluated diverse AI approaches—including CNNs, transfer learning, multi-scale feature fusion, and image augmentation—and conducted the first quantitative performance comparison between AI systems and taxonomic experts. Contribution/Results: Multiple participating systems achieved human-expert-level accuracy—or surpassed it on specific subsets—in the 10,000-class identification task, notably improving recognition rates for rare and endangered species. The challenge demonstrated the feasibility and practical potential of AI-driven, in-situ biodiversity monitoring.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Humans and AI: Voting
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Automated identification of plants has improved considerably thanks to the recent progress in deep learning and the availability of training data. However, this profusion of data only concerns a few tens of thousands of species, while the planet has nearly 369K. The LifeCLEF 2019 Plant Identification challenge (or "PlantCLEF 2019") was designed to evaluate automated identification on the flora of data deficient regions. It is based on a dataset of 10K species mainly focused on the Guiana shield and the Northern Amazon rainforest, an area known to have one of the greatest diversity of plants and animals in the world. As in the previous edition, a comparison of the performance of the systems evaluated with the best tropical flora experts was carried out. This paper presents the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Problem

• Automated plant identification faces data scarcity for most species worldwide
• Evaluating AI systems on tropical flora from data-deficient Amazon regions
• Comparing machine performance with expert botanists in species recognition

Innovation

• Deep learning for plant identification
• Focus on data-deficient tropical regions
• Comparison with expert botanist performance