Deep network for simultaneous decomposition and classification in UWB-SAR imagery

Tiep H. Vu, Lam Nguyen, Tiantong Guo, Vishal Monga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Classifying buried and obscured targets of interest from other natural and manmade clutter objects in the scene is an important problem for the U.S. Army. Targets of interest are often represented by signals captured using low-frequency (UHF to L-band) ultra-wideband (UWB) synthetic aperture radar (SAR) technology. This technology has been used in various applications, including ground penetration and sensing-through-the-wall. However, the technology still faces a significant issue regarding low-resolution SAR imagery in this particular frequency band, low radar cross sections (RCS), small objects compared to radar signal wavelengths, and heavy interference. The classification problem has been firstly, and partially, addressed by sparse representation-based classification (SRC) method which can extract noise from signals and exploit the cross-channel information. Despite providing potential results, SRC-related methods have drawbacks in representing nonlinear relations and dealing with larger training sets. In this paper, we propose a Simultaneous Decomposition and Classification Network (SDCN) to alleviate noise inferences and enhance classification accuracy. The network contains two jointly trained sub-networks: the decomposition sub-network handles denoising, while the classification sub-network discriminates targets from confusers. Experimental results show significant improvements over a network without decomposition and SRC-related methods.

Original languageEnglish (US)
Title of host publication2018 IEEE Radar Conference, RadarConf 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages553-558
Number of pages6
ISBN (Electronic)9781538641675
DOIs
StatePublished - Jun 8 2018
Event2018 IEEE Radar Conference, RadarConf 2018 - Oklahoma City, United States
Duration: Apr 23 2018Apr 27 2018

Publication series

Name2018 IEEE Radar Conference, RadarConf 2018

Other

Other2018 IEEE Radar Conference, RadarConf 2018
CountryUnited States
CityOklahoma City
Period4/23/184/27/18

Fingerprint

radar imagery
synthetic aperture radar
Synthetic aperture radar
Ultra-wideband (UWB)
broadband
Decomposition
decomposition
radar cross sections
Radar cross section
clutter
ultrahigh frequencies
classifying
inference
Frequency bands
radar
Radar
education
penetration
low frequencies
interference

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Signal Processing
  • Instrumentation

Cite this

Vu, T. H., Nguyen, L., Guo, T., & Monga, V. (2018). Deep network for simultaneous decomposition and classification in UWB-SAR imagery. In 2018 IEEE Radar Conference, RadarConf 2018 (pp. 553-558). (2018 IEEE Radar Conference, RadarConf 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RADAR.2018.8378619
Vu, Tiep H. ; Nguyen, Lam ; Guo, Tiantong ; Monga, Vishal. / Deep network for simultaneous decomposition and classification in UWB-SAR imagery. 2018 IEEE Radar Conference, RadarConf 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 553-558 (2018 IEEE Radar Conference, RadarConf 2018).
@inproceedings{a1ebd35c89444a0a8f69b1f9f7a5436e,
title = "Deep network for simultaneous decomposition and classification in UWB-SAR imagery",
abstract = "Classifying buried and obscured targets of interest from other natural and manmade clutter objects in the scene is an important problem for the U.S. Army. Targets of interest are often represented by signals captured using low-frequency (UHF to L-band) ultra-wideband (UWB) synthetic aperture radar (SAR) technology. This technology has been used in various applications, including ground penetration and sensing-through-the-wall. However, the technology still faces a significant issue regarding low-resolution SAR imagery in this particular frequency band, low radar cross sections (RCS), small objects compared to radar signal wavelengths, and heavy interference. The classification problem has been firstly, and partially, addressed by sparse representation-based classification (SRC) method which can extract noise from signals and exploit the cross-channel information. Despite providing potential results, SRC-related methods have drawbacks in representing nonlinear relations and dealing with larger training sets. In this paper, we propose a Simultaneous Decomposition and Classification Network (SDCN) to alleviate noise inferences and enhance classification accuracy. The network contains two jointly trained sub-networks: the decomposition sub-network handles denoising, while the classification sub-network discriminates targets from confusers. Experimental results show significant improvements over a network without decomposition and SRC-related methods.",
author = "Vu, {Tiep H.} and Lam Nguyen and Tiantong Guo and Vishal Monga",
year = "2018",
month = "6",
day = "8",
doi = "10.1109/RADAR.2018.8378619",
language = "English (US)",
series = "2018 IEEE Radar Conference, RadarConf 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "553--558",
booktitle = "2018 IEEE Radar Conference, RadarConf 2018",
address = "United States",

}

Vu, TH, Nguyen, L, Guo, T & Monga, V 2018, Deep network for simultaneous decomposition and classification in UWB-SAR imagery. in 2018 IEEE Radar Conference, RadarConf 2018. 2018 IEEE Radar Conference, RadarConf 2018, Institute of Electrical and Electronics Engineers Inc., pp. 553-558, 2018 IEEE Radar Conference, RadarConf 2018, Oklahoma City, United States, 4/23/18. https://doi.org/10.1109/RADAR.2018.8378619

Deep network for simultaneous decomposition and classification in UWB-SAR imagery. / Vu, Tiep H.; Nguyen, Lam; Guo, Tiantong; Monga, Vishal.

2018 IEEE Radar Conference, RadarConf 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 553-558 (2018 IEEE Radar Conference, RadarConf 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Deep network for simultaneous decomposition and classification in UWB-SAR imagery

AU - Vu, Tiep H.

AU - Nguyen, Lam

AU - Guo, Tiantong

AU - Monga, Vishal

PY - 2018/6/8

Y1 - 2018/6/8

N2 - Classifying buried and obscured targets of interest from other natural and manmade clutter objects in the scene is an important problem for the U.S. Army. Targets of interest are often represented by signals captured using low-frequency (UHF to L-band) ultra-wideband (UWB) synthetic aperture radar (SAR) technology. This technology has been used in various applications, including ground penetration and sensing-through-the-wall. However, the technology still faces a significant issue regarding low-resolution SAR imagery in this particular frequency band, low radar cross sections (RCS), small objects compared to radar signal wavelengths, and heavy interference. The classification problem has been firstly, and partially, addressed by sparse representation-based classification (SRC) method which can extract noise from signals and exploit the cross-channel information. Despite providing potential results, SRC-related methods have drawbacks in representing nonlinear relations and dealing with larger training sets. In this paper, we propose a Simultaneous Decomposition and Classification Network (SDCN) to alleviate noise inferences and enhance classification accuracy. The network contains two jointly trained sub-networks: the decomposition sub-network handles denoising, while the classification sub-network discriminates targets from confusers. Experimental results show significant improvements over a network without decomposition and SRC-related methods.

AB - Classifying buried and obscured targets of interest from other natural and manmade clutter objects in the scene is an important problem for the U.S. Army. Targets of interest are often represented by signals captured using low-frequency (UHF to L-band) ultra-wideband (UWB) synthetic aperture radar (SAR) technology. This technology has been used in various applications, including ground penetration and sensing-through-the-wall. However, the technology still faces a significant issue regarding low-resolution SAR imagery in this particular frequency band, low radar cross sections (RCS), small objects compared to radar signal wavelengths, and heavy interference. The classification problem has been firstly, and partially, addressed by sparse representation-based classification (SRC) method which can extract noise from signals and exploit the cross-channel information. Despite providing potential results, SRC-related methods have drawbacks in representing nonlinear relations and dealing with larger training sets. In this paper, we propose a Simultaneous Decomposition and Classification Network (SDCN) to alleviate noise inferences and enhance classification accuracy. The network contains two jointly trained sub-networks: the decomposition sub-network handles denoising, while the classification sub-network discriminates targets from confusers. Experimental results show significant improvements over a network without decomposition and SRC-related methods.

UR - http://www.scopus.com/inward/record.url?scp=85049980012&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049980012&partnerID=8YFLogxK

U2 - 10.1109/RADAR.2018.8378619

DO - 10.1109/RADAR.2018.8378619

M3 - Conference contribution

AN - SCOPUS:85049980012

T3 - 2018 IEEE Radar Conference, RadarConf 2018

SP - 553

EP - 558

BT - 2018 IEEE Radar Conference, RadarConf 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Vu TH, Nguyen L, Guo T, Monga V. Deep network for simultaneous decomposition and classification in UWB-SAR imagery. In 2018 IEEE Radar Conference, RadarConf 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 553-558. (2018 IEEE Radar Conference, RadarConf 2018). https://doi.org/10.1109/RADAR.2018.8378619