Biodiversity Preservation

AI supports biodiversity preservation by monitoring living assets, flagging risk, and turning scattered observations into actionable conservation data.

Business impact

• Ancient and notable trees covered by the project — Ping An reports coverage of more than 55,000 trees.
• Risk protection provided by the project — Ping An reports over RMB 700 million in risk protection.
• Tree-condition sensor monitoring — Ping An monitors tree conditions 24/7.

Data requirements

• Environmental sensor data (sensor) — Ping An monitors soil moisture, air quality, and tree health.
• Remote sensing data (image) — Ping An uses remote sensing in EagleX to track hazards.
• Meteorological monitoring data (tabular) — Ping An tracks weather-related hazards such as extreme weather, floods, and typhoons.
• Online photographs (image) — sharkPulse scans images from online sources for shark sightings.
• Image metadata and contextual clues (text) — The shark system extracts location and timestamp information and preliminary species identifications.

AI methods and techniques

• Anomaly detection — Ping An uses AI analytics to detect major weather and hazard risks for ancient trees.
• Computer vision — Virginia Tech uses convolutional neural networks to identify shark images online.
• Information extraction — The sharkPulse system extracts location, timestamp, and preliminary species data from images.
• Validation workflow — Public reviewers and experts confirm shark identifications before records enter the database.

AI models and model families

convolutional neural networks

Industries

Real-world evidence

2 documented case studies on record.

Companies using this: Ping An Insurance (Group) Company of China, Ltd., Virginia Tech.