On distinguishing between human individuals in micro-Doppler signatures
2013 (English)In: 14th International Radar Symposium (IRS), 2013, 865-870 p.Conference paper (Refereed)
Radar micro-Doppler signatures (MDS) of humans are created by movements of body parts, such as legs and arms. MDSs can be used in security applications to detect humans and classify their type and activity. Target association and tracking, which can facilitate the classification, become easier if it is possible to distinguish between human individuals by their MDSs. By this we mean to recognize the same individual in a short time frame but not to establish the identity of the individual. In this paper we perform a statistical experiment in which six test persons are able to distinguish between walking human individuals from their MDSs. From this we conclude that there is information in the MDSs of the humans to distinguish between different individuals, which also can be used by a machine. Based on the results of the best test persons we also discuss features in the MDSs that could be utilized to make this processing possible.
Place, publisher, year, edition, pages
2013. 865-870 p.
Doppler radar, image classification, national security, object detection, object recognition, radar imaging, statistical analysis, target tracking
Control Engineering Signal Processing
IdentifiersURN: urn:nbn:se:liu:diva-98190ISBN: 978-1-4673-4821-8OAI: oai:DiVA.org:liu-98190DiVA: diva2:652529
14th International Radar Symposium (IRS), Dresden, Germany, June 19 – 21, 2013