Discovery and analysis of topographic features using learning algorithms: A seamount case study. Issue 12 (20th June 2013)

• Record Type: Journal Article
• Authors: Valentine, Andrew P.; Kalnins, Lara M.; Trampert, Jeannot
• Is Part Of: Geophysical research letters. Volume 40:Issue 12(2013:Jun.)
• Page Start: 3048
• Page End: 3054
• Publication Date: 2013-06-20
• Subjects: geomorphology -- seamounts -- neural networks -- landforms -- automatic identification; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1002/grl.50615
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract : [1] Identifying and cataloging occurrences of particular topographic features are important but time‐consuming tasks. Typically, automation is challenging, as simple models do not fully describe the complexities of natural features. We propose a new approach, where a particular class of neural network (the "autoencoder") is used to assimilate the characteristics of the feature to be cataloged, and then applied to a systematic search for new examples. To demonstrate the feasibility of this method, we construct a network that may be used to find seamounts in global bathymetric data. We show results for two test regions, which compare favorably with results from traditional algorithms. Key Points: Neural networks can learn complex features in a hand‐selected set of landforms They can then be used to systematically search for further examples We demonstrate the method by identifying seamounts in bathymetric data