We're delighted to share that we have successfully trained an AI-model to recognise different water recreation users.
Developed in conjunction with Brunel University, our new AI image recognition system can recognise 20 different recreational watercraft including dinghies, jet skis, kayaks, rowers and stand-up paddle boarders.
The newly trained and adapted system can collect, process and identify recreational watercraft in real-time making understanding the use of waterways more efficient and less costly.
NASH Maritime Director Ed Rogers, said,
“We believe our AI image recognition platform brings a step change in how non-AIS marine traffic can be measured and monitored. And, from our perspective, increases in the accuracy and efficiency of preparing Navigation Risk Assessments (NRA) where water recreation users are concerned.
“Integral to NRA is the analysis of vessel traffic data. Until now there was no easy way to collect and analyse recreational craft movements. Engaging with local stakeholders was the only way to get an idea of what was happening on the water which was then ground-truthed with observers. The former is not particularly robust, whilst direct observation is resource intensive, time-limited and expensive.
“Now, our readily deployed AI system can take a CCTV feed, process, identify and record different types of recreational craft in real-time with 95%accuracy.”
We were awarded an Innovate UK grant through its ‘Bridge-AI’ scheme to enhance our existing AI image recognition platform.
As well as using the system in undertaking navigation risk assessments, we're now engaging with other parties who have an interest in better understanding and managing water recreation use.
