Saved in:
Bibliographic Details
Main Authors: Day, Sarah, Dimino, Jesse, Jester, Matt, Keegan, Kaitlin, Weighill, Thomas
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.16150
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915253773664256
author Day, Sarah
Dimino, Jesse
Jester, Matt
Keegan, Kaitlin
Weighill, Thomas
author_facet Day, Sarah
Dimino, Jesse
Jester, Matt
Keegan, Kaitlin
Weighill, Thomas
contents In this paper we evaluate the performance of topological features for generalizable and robust classification of firn image data, with the broader goal of understanding the advantages, pitfalls, and trade-offs in topological featurization. Firn refers to layers of granular snow within glaciers that haven't been compressed into ice. This compactification process imposes distinct topological and geometric structure on firn that varies with depth within the firn column, making topological data analysis (TDA) a natural choice for understanding the connection between depth and structure. We use two classes of topological features, sublevel set features and distance transform features, together with persistence curves, to predict sample depth from microCT images. A range of challenging training-test scenarios reveals that no one choice of method dominates in all categories, and uncoveres a web of trade-offs between accuracy, interpretability, and generalizability.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16150
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Classification of Firn Data via Topological Features
Day, Sarah
Dimino, Jesse
Jester, Matt
Keegan, Kaitlin
Weighill, Thomas
Computer Vision and Pattern Recognition
Algebraic Topology
55N31, 68T45
In this paper we evaluate the performance of topological features for generalizable and robust classification of firn image data, with the broader goal of understanding the advantages, pitfalls, and trade-offs in topological featurization. Firn refers to layers of granular snow within glaciers that haven't been compressed into ice. This compactification process imposes distinct topological and geometric structure on firn that varies with depth within the firn column, making topological data analysis (TDA) a natural choice for understanding the connection between depth and structure. We use two classes of topological features, sublevel set features and distance transform features, together with persistence curves, to predict sample depth from microCT images. A range of challenging training-test scenarios reveals that no one choice of method dominates in all categories, and uncoveres a web of trade-offs between accuracy, interpretability, and generalizability.
title Classification of Firn Data via Topological Features
topic Computer Vision and Pattern Recognition
Algebraic Topology
55N31, 68T45
url https://arxiv.org/abs/2504.16150