DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging

Rawin Assabumrungrat; Nakorn Kumchaiseemak; Jianping Wang; Dingyang Wang; Phoom Punpeng; Francesco Fioranelli; Theerawit Wilaiprasitporn

doi:10.1109/SENSORS56945.2023.10325198

DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging

Rawin Assabumrungrat
, Nakorn Kumchaiseemak
, Jianping Wang
, Dingyang Wang
, Phoom Punpeng
, Francesco Fioranelli^*
, Theerawit Wilaiprasitporn

^*Corresponding author for this work

Tohoku University
Vidyasirimedhi Institute of Science and Technology
Delft University of Technology
Ruamrudee International School

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

We present a deep learning-based approach called DipSAR for reconstructing millimeter-wave synthetic aperture radar (SAR) images from sparse samples. The primary challenge lies in the requirement of a large training dataset for deep learning schemes. To overcome this issue, we employ the deep image prior (DIP) technique, which eliminates the need for a large dataset and instead utilizes only the sparse sample itself. Our proposed DipSAR model recovers missing samples from sparse data and reconstructs the SAR image using a conventional method. In this study, we utilize an existing SAR dataset and create fourteen different patterns to generate additional sparse samples by removing certain data points. We then evaluate the performance of DipSAR in comparison to the conventional method. The results show that DipSAR outperforms the conventional method in terms of the intersection over union (IoU) score.

Original language	English
Title of host publication	2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350303872
DOIs	https://doi.org/10.1109/SENSORS56945.2023.10325198
Publication status	Published - 2023
Externally published	Yes
Event	2023 IEEE SENSORS, SENSORS 2023 - Vienna, Austria Duration: 29 Oct 2023 → 1 Nov 2023

Publication series

Name	Proceedings of IEEE Sensors
ISSN (Print)	1930-0395
ISSN (Electronic)	2168-9229

Conference

Conference	2023 IEEE SENSORS, SENSORS 2023
Country/Territory	Austria
City	Vienna
Period	29/10/23 → 1/11/23

Keywords

deep image prior
millimeter-wave
near-field imaging
sparse data
synthetic aperture radar

Access to Document

10.1109/SENSORS56945.2023.10325198

Cite this

Assabumrungrat, R., Kumchaiseemak, N., Wang, J., Wang, D., Punpeng, P., Fioranelli, F., & Wilaiprasitporn, T. (2023). DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging. In 2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings (Proceedings of IEEE Sensors). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SENSORS56945.2023.10325198

@inproceedings{d94b7dac2bb241c3ad946fc2bf53e28e,

title = "DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging",

abstract = "We present a deep learning-based approach called DipSAR for reconstructing millimeter-wave synthetic aperture radar (SAR) images from sparse samples. The primary challenge lies in the requirement of a large training dataset for deep learning schemes. To overcome this issue, we employ the deep image prior (DIP) technique, which eliminates the need for a large dataset and instead utilizes only the sparse sample itself. Our proposed DipSAR model recovers missing samples from sparse data and reconstructs the SAR image using a conventional method. In this study, we utilize an existing SAR dataset and create fourteen different patterns to generate additional sparse samples by removing certain data points. We then evaluate the performance of DipSAR in comparison to the conventional method. The results show that DipSAR outperforms the conventional method in terms of the intersection over union (IoU) score.",

keywords = "deep image prior, millimeter-wave, near-field imaging, sparse data, synthetic aperture radar",

author = "Rawin Assabumrungrat and Nakorn Kumchaiseemak and Jianping Wang and Dingyang Wang and Phoom Punpeng and Francesco Fioranelli and Theerawit Wilaiprasitporn",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE SENSORS, SENSORS 2023 ; Conference date: 29-10-2023 Through 01-11-2023",

year = "2023",

doi = "10.1109/SENSORS56945.2023.10325198",

language = "English",

series = "Proceedings of IEEE Sensors",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings",

address = "United States",

}

Assabumrungrat, R, Kumchaiseemak, N, Wang, J, Wang, D, Punpeng, P, Fioranelli, F & Wilaiprasitporn, T 2023, DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging. in 2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings. Proceedings of IEEE Sensors, Institute of Electrical and Electronics Engineers Inc., 2023 IEEE SENSORS, SENSORS 2023, Vienna, Austria, 29/10/23. https://doi.org/10.1109/SENSORS56945.2023.10325198

DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging. / Assabumrungrat, Rawin; Kumchaiseemak, Nakorn; Wang, Jianping et al.
2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings of IEEE Sensors).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - DipSAR

T2 - 2023 IEEE SENSORS, SENSORS 2023

AU - Assabumrungrat, Rawin

AU - Kumchaiseemak, Nakorn

AU - Wang, Jianping

AU - Wang, Dingyang

AU - Punpeng, Phoom

AU - Fioranelli, Francesco

AU - Wilaiprasitporn, Theerawit

PY - 2023

Y1 - 2023

N2 - We present a deep learning-based approach called DipSAR for reconstructing millimeter-wave synthetic aperture radar (SAR) images from sparse samples. The primary challenge lies in the requirement of a large training dataset for deep learning schemes. To overcome this issue, we employ the deep image prior (DIP) technique, which eliminates the need for a large dataset and instead utilizes only the sparse sample itself. Our proposed DipSAR model recovers missing samples from sparse data and reconstructs the SAR image using a conventional method. In this study, we utilize an existing SAR dataset and create fourteen different patterns to generate additional sparse samples by removing certain data points. We then evaluate the performance of DipSAR in comparison to the conventional method. The results show that DipSAR outperforms the conventional method in terms of the intersection over union (IoU) score.

AB - We present a deep learning-based approach called DipSAR for reconstructing millimeter-wave synthetic aperture radar (SAR) images from sparse samples. The primary challenge lies in the requirement of a large training dataset for deep learning schemes. To overcome this issue, we employ the deep image prior (DIP) technique, which eliminates the need for a large dataset and instead utilizes only the sparse sample itself. Our proposed DipSAR model recovers missing samples from sparse data and reconstructs the SAR image using a conventional method. In this study, we utilize an existing SAR dataset and create fourteen different patterns to generate additional sparse samples by removing certain data points. We then evaluate the performance of DipSAR in comparison to the conventional method. The results show that DipSAR outperforms the conventional method in terms of the intersection over union (IoU) score.

KW - deep image prior

KW - millimeter-wave

KW - near-field imaging

KW - sparse data

KW - synthetic aperture radar

UR - https://www.scopus.com/pages/publications/85179756621

U2 - 10.1109/SENSORS56945.2023.10325198

DO - 10.1109/SENSORS56945.2023.10325198

M3 - Conference contribution

AN - SCOPUS:85179756621

T3 - Proceedings of IEEE Sensors

BT - 2023 IEEE SENSORS, SENSORS 2023 - Conference Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 29 October 2023 through 1 November 2023

ER -

DipSAR: Deep Image Prior for Sparse Sampled Near-Field SAR Millimeter-Wave Imaging

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this