Multibeam Data Mining : Automated Detection of Seafloor Anomalies and Seamounts to Enhance Understanding of Seafloor Processes
This dissertation demonstrates that combining ensemble-based quality control with machine-learning–driven feature detection enables robust, scalable, and auditable analysis of large multibeam echosounder bathymetry datasets. By substantially reducing operator-dependent intervention while preserving expert traceability, the proposed methods improve consistency across surveys, regions, and sensors and facilitate the discovery of previously unresolved geomorphic features. Beyond seafloor mapping, the presented design principles are broadly transferable to other large-scale geospatial raster analyses. Future developments integrating backscatter information and weakly supervised learning promise further gains in interpretability, coverage, and efficiency, supporting next-generation, data-driven seafloor exploration.
Preview
Rights
Use and reproduction:
Please note that individual components of the publication may be subject to other licensing or copyright conditions.