Urgent.News

What's breaking now, across thousands of outlets.

AI

AI helps researchers track beluga whales using aerial drone images

For Arctic researchers, counting beluga whales from the air can mean painstakingly examining thousands of aerial images—searching for animals that may be difficult to see and even harder to distinguish from one another. The challenge is especially significant when researchers are trying to identify calves, whose smaller size, lower numbers and variable appearance can make them easy to miss.

AI helps researchers track beluga whales using aerial drone images

Researchers from Florida Atlantic University and Fisheries and Oceans Canada have developed an artificial intelligence system to streamline the process of tracking beluga whales using aerial drone images. Traditionally, counting and identifying beluga whales from aerial images has been a labor-intensive task, requiring thousands of manual annotations.

The new semi-supervised AI model, SEMI-DETR, requires only a small fraction of labeled data—just 1% of the available annotated images—to achieve comparable accuracy to conventional supervised systems. At the 1% labeling level, SEMI-DETR demonstrated a 24.3% improvement in detecting beluga calves (younger whales) compared to the supervised baseline, while only delivering a 16.8% improvement for adult whales.

The AI system achieved a mean average precision of 19.2% using just 34 labeled images, representing a 20% relative improvement over the best-performing supervised model. This breakthrough offers conservation scientists a more efficient method to monitor beluga populations while minimizing the manual annotation effort required to train AI models for wildlife detection.

Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at phys.org →

More in AI

More from Thursday 19 November →