New technology gives sea safety hopes
The general application of applying transfer learning. First, a CNN (MobileNet) is trained on an extremely large dataset. Next, the CNN is fine-tuned on a smaller dataset. The architecture of a CNN consists of convolutional layers (blue), which have a series of convolutional kernels that extract features from the images. The maximum pooling (red) layers help reduce the dimensionality of the data and discard redundant features. Finally, the fully connected layers (green) further reduce the dimensionality of the data into a compressed latent representation. The final layer makes a prediction into two classes: rip current or no rip current. Credit: Remote Sensing (2022). DOI: 10.3390/rs14236048

Real-time updates of rip currents could be given by a new technology that could make the beach safer.

There are segments of water that travel away from the shore. The fastest Olympian can reach speeds of 2.5 meters per second.

Artificial intelligence and deep learning have been used to develop a state-of-the-art rip current identification tool. The tool was able to detect rip currents in video and images. There is a work in the journal.

The scientists hope the technology will eventually be used by beachgoers to alert them of rip currents.

We hope that by using cameras and drones at the beach, the tool will be able to detect and warn people of rip currents. Surf lifeguards can't identify rip depending on the beach and environment. Surf lifeguards can use all the extra help they can get because some beaches are large.

New technology gives sea safety hopes
The "warmer" colors are where the AI is focusing—it gives a probability score that shows, with various confidence, that a potentially deadly rip current is in that area. Credit: NIWA

The technology was developed by feeding millions of coastal aerial images into an artificial intelligence model and teaching it to identify where rip occur in real-time regardless of weather conditions and camera angle.

SLS New Zealand's Special Projects Manager, Adam Wooler, says once the technology is operationalized, it will be useful for people to enjoy our beautiful coastline.

We want to build even more effective, accurate and reliable safety tools for New Zealand beaches. We had 90 people sadly drown in New Zealand last year, 25 of them on beaches, so we're hoping that this technology will help to reduce that number.

According to reports, rip currents are the most dangerous safety risk on the beach. They have killed more people on Australian territory than all the other disasters combined.

There is more information about Interpretable Deep Learning Applied to Rip Current Detection. There is a DOI called 10.3390.