Thermal Cameras and AI Help Ships Steer Clear of Gray Whales

2026-05-26 · Source: IEEE Spectrum · Field: Science & Research — Environmental Science & Earth Systems, Life Sciences & Biology, Artificial Intelligence & Machine Learning · Depth: Intermediate, short

Summary

A new AI-based whale detection system, developed by WhaleSpotter, launched on May 19 in San Francisco Bay to mitigate ship strikes on gray whales. This system utilizes thermal cameras positioned at Point Blunt on Angel Island and soon on a ferry, feeding footage to an AI model that identifies whale spouts. Human experts verify detections to prevent false alarms before alerts are sent to nearby ships via the U.S. Coast Guard's Vessel Traffic Service, prompting them to slow or reroute. This initiative addresses a critical issue: a study estimates an 18 percent mortality rate for gray whales entering the bay in 2026, with 40 percent of 2025's record 21 deaths attributed to ship strikes. Gray whales have increasingly detoured into the bay since 2018, possibly due to climate change affecting their Arctic food sources. The system, which logged 6,600 detections in its first 1.5 weeks, is the first to combine land-based and vessel-based monitoring for continuous coverage.

Key takeaway

For shipping operations managers navigating busy coastal waters, this AI-based thermal camera system offers a proven method to significantly reduce whale strike risks. You should consider integrating similar continuous monitoring solutions, especially in known migration corridors or newly identified whale feeding grounds. Implementing human-verified AI alerts can enhance safety and compliance, protecting marine life while maintaining operational efficiency. Explore how feeding verification data back into your AI models can further refine detection accuracy.

Key insights

AI-powered thermal camera systems can effectively detect marine mammals, reducing ship strike risks and aiding conservation efforts.

Principles

• Continuous monitoring improves detection rates.
• Human verification is crucial for AI system reliability.
• Data feedback loops enhance AI algorithm performance.

Method

The system uses thermal cameras to capture whale spouts, an AI model processes footage for detection, human experts verify, and alerts are sent to ships via Vessel Traffic Service.

In practice

• Deploy thermal cameras for 24/7 marine mammal monitoring.
• Integrate human-in-the-loop verification for critical alerts.
• Feed false positive data back to AI models for refinement.

Topics

• Whale Detection
• Thermal Cameras
• AI Monitoring
• Ship Strike Prevention
• Gray Whale Migration
• Marine Conservation
• Vessel Traffic Service

Best for: Computer Vision Engineer, AI Scientist, Research Scientist, AI Engineer

Related on AIssential

• San Francisco turns to AI to save whales from ship strikes as deaths soar — AI-powered detection and ropeless fishing gear offer solutions to protect whales from ship strikes and entanglement.
• Interest in Spoor’s bird-monitoring AI software is soaring — Computer vision effectively monitors bird populations to reduce wind turbine impact and inform environmental compliance.
• Strait of Hormuz crisis drives demand for commercial geospatial intelligence — Geospatial intelligence platforms combining diverse open-source data are crucial for real-time maritime domain awareness in contested zones.
• Researchers are combining drones and AI to make removing land mines faster and safer — Drone-based multisensor AI improves land mine detection speed, accuracy, and safety by fusing diverse data and quantifying uncertainty.
• Drownings are surging in Australia. AI can help — AI and community data collection enhance water safety by improving incident detection and lifeguard support.
• Silicon Valley bets $200M on AI data centers floating in the ocean — Ocean-based AI data centers powered by wave energy offer a novel solution to land-based infrastructure challenges.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by IEEE Spectrum.