Abstract
Automatic Target Recognition (ATR) is a valuable application of computer vision that traditionally requires copious and tedious labeling through supervised learning. This research explored if ATR can be performed on satellite imagery at a comparable accuracy to a fully supervised baseline model with a considerably smaller subset of data labelled, on the order of 10%, using a recently developed semi-supervised technique, contrastive learning. Supervised contrastive loss was explored and compared to traditional cross entropy loss. Supervised contrastive loss was found to perform significantly better with a subset of the data labelled on the XView dataset, a publicly available dataset of satellite imagery captured with .3 meter ground sampling. The caveats when nothing and everything is labelled were additionally explored. [1].
Original language | English |
---|---|
Title of host publication | Automatic Target Recognition XXXII |
Editors | Riad I. Hammoud, Timothy L. Overman, Abhijit Mahalanobis, Kristen Jaskie |
ISBN (Electronic) | 9781510650688 |
DOIs | |
State | Published - 2022 |
Event | Automatic Target Recognition XXXII 2022 - Virtual, Online Duration: Jun 6 2022 → Jun 12 2022 |
Publication series
Name | Proceedings of SPIE - The International Society for Optical Engineering |
---|---|
Volume | 12096 |
ISSN (Print) | 0277-786X |
ISSN (Electronic) | 1996-756X |
Conference
Conference | Automatic Target Recognition XXXII 2022 |
---|---|
City | Virtual, Online |
Period | 6/6/22 → 6/12/22 |
Bibliographical note
Publisher Copyright:© 2022 SPIE. All rights reserved.
Keywords
- ATR
- Classification
- Computer Vision
- Contrastive Learning
- Satellite Imagery
- Self-Supervised Learning
- Semi-Supervised Learning
ASJC Scopus subject areas
- Electronic, Optical and Magnetic Materials
- Condensed Matter Physics
- Computer Science Applications
- Applied Mathematics
- Electrical and Electronic Engineering