Machine Learning How to Sea: Recognition and classification of subsea structures using artificial intelligence and high-performance computing

Raska Soemantoro, Mario Gaimann

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Understanding subsea reliefs is typically a task given to specialist geologists
manually labelling each area. With artificial intelligence and high performance computing, we propose a method to automate this process entirely. In this project, we developed an AI-based method to automate the classification of subsea reliefs. Our method comprises 5 steps; pre-processing of geological data, performing selective search, training two R-CNN models, and evaluating the predictions.
Original languageEnglish
Title of host publicationSummer of HPC 2021 Reports
Place of PublicationBrussels, Belgium
PublisherPRACE: Partnership for Advanced Computing in Europe
Pages35-37
Number of pages3
Publication statusPublished - 2022

Fingerprint

Dive into the research topics of 'Machine Learning How to Sea: Recognition and classification of subsea structures using artificial intelligence and high-performance computing'. Together they form a unique fingerprint.

Cite this