TY - GEN
T1 - Analysis of Hyperspectral Data by Means of Transport Models and Machine Learning
AU - Czaja, Wojciech
AU - Dong, Dong
AU - Jabin, Pierre Emmanuel
AU - Njeunje, Franck O.Ndjakou
N1 - Funding Information:
WC was partially supported by LTS grant D00030014 and by NSF grant DMS-1738003. DD was partially supported by LTS grant DO 0052. PEJ was partially supported by NSF Grant 161453, NSF Grant RNMS (Ki-Net) 1107444 and by LTS grant D00030014.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/9/26
Y1 - 2020/9/26
N2 - We present a new physics-inspired method for analysis of hyperspectral imagery (HSI). The method is based on the concept of transport models for graphs. The proposed approach generalizes existing dimension reduction and feature extraction algorithms, by replacing the role of diffusion processes, as a measure of estimating proximity, with dynamical systems. This approach allows us to exploit different and new relationships within the complex data structures, such as those arising in HSI. We demonstrate this by proposing a specific multi-scale algorithm in which transport models are used to translate the information about contextual similarities of material classes to enhance feature extraction and classification results. This point is illustrated with a series of computational experiments.
AB - We present a new physics-inspired method for analysis of hyperspectral imagery (HSI). The method is based on the concept of transport models for graphs. The proposed approach generalizes existing dimension reduction and feature extraction algorithms, by replacing the role of diffusion processes, as a measure of estimating proximity, with dynamical systems. This approach allows us to exploit different and new relationships within the complex data structures, such as those arising in HSI. We demonstrate this by proposing a specific multi-scale algorithm in which transport models are used to translate the information about contextual similarities of material classes to enhance feature extraction and classification results. This point is illustrated with a series of computational experiments.
UR - http://www.scopus.com/inward/record.url?scp=85101959757&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101959757&partnerID=8YFLogxK
U2 - 10.1109/IGARSS39084.2020.9323215
DO - 10.1109/IGARSS39084.2020.9323215
M3 - Conference contribution
AN - SCOPUS:85101959757
T3 - International Geoscience and Remote Sensing Symposium (IGARSS)
SP - 3680
EP - 3683
BT - 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020
Y2 - 26 September 2020 through 2 October 2020
ER -