A groundbreaking study published in Ecology and Evolution reveals how a computer algorithm can accurately distinguish individual beavers by analyzing the unique patterns of scales on their tails, akin to human fingerprints.
Beavers, renowned for their leathery tails used for steering while swimming and signaling alarm, possess tail scales that serve as distinctive identifiers. The study, focusing on Eurasian beavers (Castor fiber), marks a significant development in wildlife monitoring and conservation efforts.
Traditionally, population estimation methods for beavers involved invasive techniques such as ear tagging and radio collaring, which can induce stress in the animals. In response, researchers turned to pattern-learning artificial intelligence, training it on images of tails from 100 hunted Eurasian beavers in Norway. Remarkably, the AI achieved a remarkable 96% accuracy in distinguishing between individual beavers based on tail scale patterns.
The potential applications of this technology extend to fieldwork, as demonstrated in a previous study by the same team. By equipping automatic cameras with small plastic lenses, researchers were able to capture high-quality images in the wild, facilitating computer-based identification of beavers. This innovative approach promises to revolutionize wildlife monitoring efforts, offering a non-invasive and efficient means of tracking populations and aiding conservation initiatives.
More: https://www.science.org/content/article/ai-tells-beavers-apart-fingerprint-patterns-their-tails
