With autonomous underwater vehicles (AUVs) and diving robots exploring the deep seas, the amount of ocean data is quickly growing.

“Technological advances like 4K cameras, autonomous robots, high-capacity batteries, and LED lighting now allow systematic optical monitoring at large spatial scale and shorter time, but with increased data volume and velocity,” according to the Scientific Data research article.

However, Dr. Timm Schoening from the “Deep Sea Monitoring” working group headed by Prof. Dr. Jens Greinert at GEOMAR, has developed a solution to upgrade image analysis.

“Over the past 3 years, we have developed a standardized workflow that makes it possible to scientifically evaluate large amounts of image data systematically and sustainably,” says Schoening.

The inspiration behind this data management system all started with a JPI Oceans project called “MiningImpact, which used the ABYSS AUV to evaluate the prolonged effects of manganese nodule mining within a deep-sea environment. The AUV was fashioned with a digital camera system, and its data was filtered through the new workflow system.

The process is broken down into three main steps:

• Data actuation
• Data curation
• Data management

During each step, important factors are considered, including camera setup, optimal lightning parameters, and the type of measurements needed for the intended experiment. In ABYSS’ case, the AUV auto-recorded its position, dive depth, and water properties.

“For data processing, it is essential to link the camera’s image data with the diving robot’s metadata,” says Schoening. “All this information has to be linked to the respective image because it provides important information for subsequent evaluation.”

After 30 dives, ABYSS’ gathered more than 500,000 seafloor images. The new procure automatically removes unusable image data (motion blur), and guarantees all the information is safely and efficiently corralled.

“Until then, however, a large number of time-consuming steps had been necessary,” says Schoening. “Now the method can be transferred to any project, even with other AUVs or camera systems.”

Artificial intelligence (AI) was the final piece of the puzzle, which analyzed whether ABYSS’ photo contained a manganese nodule, along with its size and position.

During next year’s nodule research using the improved data management system, the team hopes to conduct on-board image material evaluation.

“Therefore we will take some particularly powerful computers with us on board,” says Schoening.

To learn more, read the article, “An acquisition, curation and management workflow for sustainable, terabyte-scale marine image analysis,” published in Scientific Data.