The sensor is designed to monitor zooplankton and phytoplankton that displays positive phototaxis and are attracted to the sensor. Plankton serve as critical indicators of marine ecosystem health, and their monitoring provides valuable early warnings of ecological changes.

The sensor utilizes 420 nm light to attract zooplankton.

Resolution is around 3 micrometers, which means that we are able to detect and classify zooplankton and phytoplankton, that display positive phototaxis, down to 100 micrometers. The upper limit for target organisms we can detect is around 3000 micrometers.

We train a compact inference model based on the auto clustered data, using a vision transformer for YOLO / Detectron2 object detection.

The submerged microscope uses a single cable for power and gigabit ethernet. Currently, all image and video data are uploaded to the shore through high-speed links. As we improve our auto-segmentation model, we'll only send the important parts of the images and videos to reduce data transfer. Eventually, our AI models will analyze the data directly on the submerged camera or an on-pen AI edge server, eliminating the need for high-speed data uploads.

Yes, FP Aqua can develop an API.

Yes, the technology can be adapted for monitoring various marine species and ecosystems beyond just salmon.

We provide crucial tools for assessing the state of marine ecosystems. By offering precise and continuous monitoring, the solution can identify early signs of ecological stress and evaluate the effectiveness of initiatives aimed at improving ocean health.

Partnerships are a critical component of our strategy. We are actively seeking additional partners for technology validation and future collaborations. We believe our technology aligns perfectly with the industry’s trend towards digitalization and partnerships.