Wildlife Monitoring

AI monitors wildlife from field data, detecting animal activity, movement, and abnormal events.

Business impact

• Coverage area monitored — One deployment covers some 40,000 square miles across 15 parks.
• Facilities or reserves covered — One deployment reports more than 40 South African reserves covered.
• Assets deployed — One deployment reports over 1,200 RhinoWatches deployed.
• Recall — One study reports recall above 70% for elephant-rumble detection.

Data requirements

• Field recordings (audio) — Used to detect animal vocalizations such as elephant rumbles.
• Animal telemetry (tabular) — Used to show movement and location within protected areas.
• Environmental data (tabular) — Used with field data to predict likely poaching locations and dates.
• Camera imagery (image) — Used in wildlife observation systems that monitor animal activity visually.

AI methods and techniques

• Machine learning — Used to analyze field, environmental, and monitoring data for wildlife signals and risk patterns.
• Classification — Used to identify animal calls or species from recorded signals and images.
• Anomaly detection — Used to alert authorities when behavior or conditions look abnormal.
• Prediction — Used to estimate likely poaching locations and times from combined data.
• Computer vision — Used in monitoring systems that analyze camera images of wildlife activity.

AI models and model families

AdaBoost

Industries

Real-world evidence

5 documented case studies on record.

Companies using this: Rouxcel Technology, Marwell Zoo, Lewa Wildlife Conservancy, Eastern Tropical Pacific Marine Conservation Corridor (CMAR), Cornell University.