liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Track-oriented self-supervised learning for micro-Doppler target classification
Norwegian Def Res Estab, Norway; Univ Oslo, Norway.
Norwegian Def Res Estab, Norway; Univ Oslo, Norway.
Linköping University, Department of Electrical Engineering, Computer Vision. Linköping University, Faculty of Science & Engineering. Univ Oslo, Norway.ORCID iD: 0000-0002-6763-5487
2024 (English)In: 2024 IEEE RADAR CONFERENCE, RADARCONF 2024, IEEE , 2024Conference paper, Published paper (Refereed)
Abstract [en]

Deep learning's shift from supervised to unsupervised methods has revolutionized many applications, exemplified by innovations such as DALL-E-2 and Chat-GPT. This paper focuses on extending the application of self-supervised learning, specifically in the radar domain, by introducing a simple, yet effective approach to training self-supervised models inspired by SimCLR, but tailored for micro-Doppler spectrograms. By using a target tracker we can, for a given track choose different contrastive samples by selecting spectrogram snippets from random time points along the whole track. This approach leverages the natural augmentation that occurs during a track sequence of a target. Despite its simplicity, this approach outperforms traditional augmentation techniques when training contrastive models. We demonstrated our approach by training a foundation model for radar classification and fine-tuning the model on two separate datasets. Our training approach improves classification performance significantly, and in a scenario where we have 100 times more unlabeled than labeled training samples, we boost performance with an impressive 24.79 percentage points (pp.), over a conventional supervised approach.

Place, publisher, year, edition, pages
IEEE , 2024.
Series
IEEE National Radar Conference Proceedings, ISSN 1097-5659, E-ISSN 2375-5318
Keywords [en]
radar; unsupervised; self-supervised; machine learning; classification; tracking; deep learning; micro-Doppler; UAV
National Category
Computer Vision and Learning Systems
Identifiers
URN: urn:nbn:se:liu:diva-209945DOI: 10.1109/RADARCONF2458775.2024.10548787ISI: 001253160700119Scopus ID: 2-s2.0-85196791896ISBN: 9798350329216 (print)ISBN: 9798350329209 (electronic)OAI: oai:DiVA.org:liu-209945DiVA, id: diva2:1915046
Conference
IEEE Radar Conference (RadarConf), Denver, CO, may 06-10, 2024
Available from: 2024-11-21 Created: 2024-11-21 Last updated: 2026-06-22

Open Access in DiVA

fulltext(4288 kB)83 downloads
File information
File name FULLTEXT01.pdfFile size 4288 kBChecksum SHA-512
fc885178d7d648d1b5f0b9cd85c72d73befd1e03a49416eedf2441b709b449a2f4a8ac90cbab17f14660570ee529567e981c1f9d52911c671d1fff5aa176096d
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Search in DiVA

By author/editor
Ahlberg, Jörgen
By organisation
Computer VisionFaculty of Science & Engineering
Computer Vision and Learning Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 83 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 102 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf