AI Revolutionizes Snow Leopard Population Counts, Boosting Conservation Efforts

The Challenge of Counting Snow Leopards

Snow leopards, known as the "ghosts of the mountains," are notoriously difficult to track and count. Their elusive nature and ability to blend seamlessly with their environment have made traditional population estimates challenging and often inaccurate. In 2017, the International Union for Conservation of Nature reclassified snow leopards from endangered to vulnerable, estimating a global population between 2,500 and 10,000 adults in the wild 1

1

Source: Popular Science

Camera Traps and Identification Challenges

To overcome the limitations of traditional tracking methods, conservationists from the Wildlife Conservation Society deployed automated camera traps in Afghanistan. These devices capture thousands of images over months, providing rare glimpses of snow leopards. However, identifying individual leopards from these images proved to be a significant challenge 2

2

Each snow leopard has a unique pattern of black rosettes on its coat, but identifying individuals by these patterns is slow, subjective, and prone to error. Factors such as odd angles, poor lighting, and partially obscured animals make accurate identification difficult. One study found that human errors led to overestimating the snow leopard population by more than 30% 1

1

AI-Powered Solutions

Researchers have turned to artificial intelligence to address these challenges. Two AI algorithms were evaluated for their effectiveness in identifying individual snow leopards:

1. HotSpotter: This algorithm identifies individual snow leopards by comparing key visual features such as coat patterns, highlighting distinctive "hot spots" with a yellow marker 3

   3

   .

2. Pose Invariant Embeddings: Similar to facial recognition technology, this method recognizes layers of abstract features in the data, identifying the same animal regardless of its position or lighting conditions 2

   2

   .

When used separately, each model achieved about 74% accuracy in correctly identifying snow leopards from a large photo library. However, when combined, the two systems achieved an impressive 85% accuracy 3

3

Source: Phys.org

Wildbook: An AI-Powered Platform for Conservation

These AI algorithms have been integrated into Wildbook, an open-source, web-based software platform developed by the nonprofit organization Wild Me and adopted by ConservationX. The combined system has been deployed on a free website called Whiskerbook, where researchers can upload images, seek matches using the algorithms, and confirm those matches with side-by-side comparisons 2

2

Related Stories

AI-Powered Acoustic Monitoring: A Game-Changer for Wildlife Conservation

AI and Vultures: A High-Tech Approach to Wildlife Conservation

AI Discovers Lions Have Two Distinct Roars, Revolutionizing Wildlife Conservation Monitoring

Generative AI is contaminating citizen science records with fake wildlife images

The Role of Human Experts

While AI has significantly improved the efficiency and accuracy of snow leopard identification, human expertise remains crucial. The AI systems quickly narrow down candidates and flag likely matches, but expert validation ensures accuracy, especially with tricky or ambiguous photos 1

1

Impact on Conservation Efforts

Source: The Conversation

The integration of AI technology in snow leopard population counts represents a significant advancement in wildlife conservation. By improving the accuracy of population estimates, researchers and conservationists can better understand the status of these elusive big cats and develop more effective strategies for their protection 3

3

This AI-powered approach not only enhances the efficiency of data processing but also contributes to more informed decision-making in conservation efforts. As part of a growing family of AI-powered wildlife platforms, Wildbook and similar technologies are helping conservation biologists work more effectively to protect endangered species and their habitats 2

2