The Korea Herald

피터빈트

Seadronix Disrupts Maritime Technology with Groundbreaking AI Model

By PRNEWSWIRE

Published : July 31, 2024 - 00:11

    • Link copied

SEOUL, South Korea, July 30, 2024 /PRNewswire/ -- Seadronix, a leader in AI-enabled autonomous ship navigation solutions, has announced the release of a groundbreaking white paper on artificial intelligence (AI) for marine object recognition. This innovative research introduces the VaDA (Vertical and Detail Attention) model, an advanced AI technology designed to recognize objects in challenging marine conditions, such as light reflections, interference, and adverse weather.

Example of benchmark dataset introduced in the white paper. Using the VaDA AI model, detailed segmentation of objects allows for accurate recognition of ships, terrain, and the sea surface.
Example of benchmark dataset introduced in the white paper. Using the VaDA AI model, detailed segmentation of objects allows for accurate recognition of ships, terrain, and the sea surface.

The VaDA model represents a significant technological breakthrough for Seadronix, the result of years of diligent work by researchers collecting and analyzing real operational data from the maritime industry, where data acquisition is notoriously difficult. VaDA's development involved an unprecedented amount of real-maritime data, achieving state-of-the-art performance in terms of both accuracy and cost-effectiveness.

In a landmark move for the marine AI industry, Seadronix has also introduced the OASIs (Ocean AI Segmentation Initiatives) dataset, the first benchmark dataset of its kind. This initiative is set to transform the industry by providing a new standard for comparing AI technologies in object recognition, addressing a critical gap in maritime AI development.

Seadronix plans to showcase its key technologies, including the VaDA model and sensor fusion, at two prestigious international conferences: the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) and the International Conference on Maritime Autonomous Surface Ships. By presenting at these events, Seadronix aims to cement its position as a global leader in maritime AI.

"The release of the OASIs benchmark dataset marks a significant milestone in the advancement of AI technology," said Hankeun Kim, Chief Technology Officer of Seadronix. "We are committed to continuously driving innovation and development in the maritime AI sector."

The introduction of object recognition AI technology is crucial for enhancing navigation assistance, safety, and ship automation. This technology is essential for autonomous navigation research, as it can detect and classify all objects at sea, from small buoys to large vessels and geographical features. Seadronix's AI capabilities set a high standard in the industry.

Seadronix's specialization in developing AI technology for marine environments, previously deemed impossible, positions the company as a leader in the digital transformation of the global shipping and port sectors. This innovation is poised to revolutionize autonomous navigation and serve as a game-changer in the industry.