🐟 AI finds ghost nets – lost fishing nets in the oceans

• Each year, 20 percent of all fishing equipment in the world's oceans is lost, resulting in approximately 50,000 tons of lost fishing nets that make up 30 percent of marine plastic waste.
• The AI-supported technology analyzes sonar images with 90 percent accuracy to mark locations where ghost nets are likely found.
• WWF Germany has manually sifted through images captured by a side-scan sonar and recovered a total of 26 tons of nets from the Baltic Sea.

WWF Germany expands search for ghost nets

WWF Germany has initiated a project to use artificial intelligence in the depths of the oceans with one mission: to track ghost nets. These abandoned fishing nets pose a significant danger to marine wildlife and ecosystems.

Lost fishing nets continue to catch fish indefinitely and become deadly traps for fish, seabirds, turtles, and marine mammals. They break down into smaller pieces and fibers over centuries, exacerbating microplastic pollution in the oceans.

Ghostnetzero.ai – an AI-supported platform

To address this problem, WWF Germany has launched the initiative ghostnetzero.ai in collaboration with Microsoft AI for Good Lab and Accenture. The platform uses AI to efficiently analyze existing sonar images specifically for ghost nets.

These sonar images are collected worldwide to secure shipping traffic or to explore locations for offshore wind turbines. When high-resolution sonar data is automatically analyzed by the AI technology, locations where ghost nets are likely to be found are marked on the platform.

Gabriele Dederer, research diver and project manager for ghost nets at WWF Germany, explains: "Ghost nets are invisible under the water surface and their detection is complex. The combination of sonar search and AI-supported detection enables a major advancement: the seabed is mapped all over the world and there is an enormous amount of data. If we can specifically check existing image data from heavily fished marine zones, this is a real change in the search for ghost nets."

Technology with 90 percent accuracy

According to WWF Germany, the AI technology is already 90 percent accurate. When a new area has been identified by the platform, a team of environmental experts checks the marked areas for validation. Depending on the nature of the seabed, it is often difficult to determine whether a suspicious structure is a sanded net or a cable.

The AI is currently being trained to reliably detect these subtle differences in sonar images from different systems. This will make it possible to evaluate existing datasets in a targeted manner.

So far, WWF Germany has manually sifted through images captured by a side-scan sonar and recovered a total of 26 tons of nets from the Baltic Sea. They are now calling on research institutes, authorities, and offshore wind power companies to donate suitable recordings via the new online platform to help advance the new technology.