An open-source image database that unlocks the power of artificial intelligence for ocean research

Dmitry Spodarets
Dmitry Spodarets

A new collaboration between MBARI and other research institutes, the open-source image database FathomNet uses state-of-the-art data processing algorithms to help process the visual data backlog. Using artificial intelligence and machine learning will remove the bottleneck in underwater image analysis and accelerate important ocean health research.

Researchers collect large amounts of visual data to observe ocean life, and machine learning is providing a way forward here. However, these approaches rely on huge datasets for training and FathomNet was created to fill that gap. It is a new web-based platform built on an API where people can upload labeled data to train new algorithms.

FathomNet addresses the lack of a standard set of existing images that could be used to train machines to recognize and catalog underwater objects and life. It combines images from a variety of sources to create a publicly available database for training underwater images.

With FathomNet, the creators aim to provide a rich, interesting benchmark to engage the machine learning community in a new field. It includes part of the MBARI database as well as National Geographic and NOAA resources. With their data as a foundation, FathomNet will help accelerate ocean research at a time when understanding the ocean is more important than ever.

Also, IBM Watson previously opened up new AI capabilities for software vendors. They announced the availability of three new AI software libraries that can be embedded directly into applications. These services include NLP, text-to-speech, and speech-to-text.