Information preserving preprocessing for improved radar target classification accuracy

H. Bozkurt; I. Erer

doi:10.1109/RAST.2017.8003014

Information preserving preprocessing for improved radar target classification accuracy

H. Bozkurt, I. Erer

Department of Electronics and Communication Engineering

Istanbul Technical University

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

1 Citation (Scopus)

Abstract

Edge preserving image decomposition is proposed as a preprocessing technique to increase the accuracy in automatic radar target classification. The radar images are decomposed trough edge preserving image decomposition methods with parameters optimized for a better classification rate. By appropriate choice of the parameters, it is possible to keep the necessary information in the residual images while transferring the redundant information to the detail planes. Thus the use of residual image with the meaningful amount of data without redundancies increases the classification performance. The proposed preprocessing scheme is validated for an experimental dataset and compared with other image decomposition methods.

Original language	English
Title of host publication	Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017
Editors	M.F. Unal, A. Hacioglu, M.S. Yildiz, O. Altan, M. Yorukoglu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	161-165
Number of pages	5
ISBN (Electronic)	9781538616031
DOIs	https://doi.org/10.1109/RAST.2017.8003014
Publication status	Published - 4 Aug 2017
Event	8th International Conference on Recent Advances in Space Technologies, RAST 2017 - Istanbul, Turkey Duration: 19 Jun 2017 → 22 Jun 2017

Publication series

Name	Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017

Conference

Conference	8th International Conference on Recent Advances in Space Technologies, RAST 2017
Country/Territory	Turkey
City	Istanbul
Period	19/06/17 → 22/06/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Funding

Data used in paper is recorded by STARLOS sensor of Sandia National Laboratory on January 1998. This data acquisition program is supported by DARPA and United States Air Force Research Laboratory under the name of Moving and Stationary Target Acquisition and Recognition (MSTAR). Data is taken by ISAR working in X band in spotlight mode. Data is in HH polarization and the reflected signal range data is used to create images. 7 targets from the dataset are used in the paper [8].

Funders	Funder number
Defense Advanced Research Projects Agency
Air Force Research Laboratory

Keywords

Bilateral Filters
component
Denoising
Image Classification
Image Processing
ISAR
Target Classification

Access to Document

10.1109/RAST.2017.8003014

Cite this

Bozkurt, H., & Erer, I. (2017). Information preserving preprocessing for improved radar target classification accuracy. In M. F. Unal, A. Hacioglu, M. S. Yildiz, O. Altan, & M. Yorukoglu (Eds.), Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017 (pp. 161-165). Article 8003014 (Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/RAST.2017.8003014

Bozkurt, H. ; Erer, I. / Information preserving preprocessing for improved radar target classification accuracy. Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017. editor / M.F. Unal ; A. Hacioglu ; M.S. Yildiz ; O. Altan ; M. Yorukoglu. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 161-165 (Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017).

@inproceedings{89eb99641a2e416cbd22118dac2ab114,

title = "Information preserving preprocessing for improved radar target classification accuracy",

abstract = "Edge preserving image decomposition is proposed as a preprocessing technique to increase the accuracy in automatic radar target classification. The radar images are decomposed trough edge preserving image decomposition methods with parameters optimized for a better classification rate. By appropriate choice of the parameters, it is possible to keep the necessary information in the residual images while transferring the redundant information to the detail planes. Thus the use of residual image with the meaningful amount of data without redundancies increases the classification performance. The proposed preprocessing scheme is validated for an experimental dataset and compared with other image decomposition methods.",

keywords = "Bilateral Filters, component, Denoising, Image Classification, Image Processing, ISAR, Target Classification",

author = "H. Bozkurt and I. Erer",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 8th International Conference on Recent Advances in Space Technologies, RAST 2017 ; Conference date: 19-06-2017 Through 22-06-2017",

year = "2017",

month = aug,

day = "4",

doi = "10.1109/RAST.2017.8003014",

language = "English",

series = "Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017",

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

pages = "161--165",

editor = "M.F. Unal and A. Hacioglu and M.S. Yildiz and O. Altan and M. Yorukoglu",

booktitle = "Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017",

address = "United States",

}

Bozkurt, H & Erer, I 2017, Information preserving preprocessing for improved radar target classification accuracy. in MF Unal, A Hacioglu, MS Yildiz, O Altan & M Yorukoglu (eds), Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017., 8003014, Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017, Institute of Electrical and Electronics Engineers Inc., pp. 161-165, 8th International Conference on Recent Advances in Space Technologies, RAST 2017, Istanbul, Turkey, 19/06/17. https://doi.org/10.1109/RAST.2017.8003014

Information preserving preprocessing for improved radar target classification accuracy. / Bozkurt, H.; Erer, I.
Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017. ed. / M.F. Unal; A. Hacioglu; M.S. Yildiz; O. Altan; M. Yorukoglu. Institute of Electrical and Electronics Engineers Inc., 2017. p. 161-165 8003014 (Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017).

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

TY - GEN

T1 - Information preserving preprocessing for improved radar target classification accuracy

AU - Bozkurt, H.

AU - Erer, I.

PY - 2017/8/4

Y1 - 2017/8/4

N2 - Edge preserving image decomposition is proposed as a preprocessing technique to increase the accuracy in automatic radar target classification. The radar images are decomposed trough edge preserving image decomposition methods with parameters optimized for a better classification rate. By appropriate choice of the parameters, it is possible to keep the necessary information in the residual images while transferring the redundant information to the detail planes. Thus the use of residual image with the meaningful amount of data without redundancies increases the classification performance. The proposed preprocessing scheme is validated for an experimental dataset and compared with other image decomposition methods.

AB - Edge preserving image decomposition is proposed as a preprocessing technique to increase the accuracy in automatic radar target classification. The radar images are decomposed trough edge preserving image decomposition methods with parameters optimized for a better classification rate. By appropriate choice of the parameters, it is possible to keep the necessary information in the residual images while transferring the redundant information to the detail planes. Thus the use of residual image with the meaningful amount of data without redundancies increases the classification performance. The proposed preprocessing scheme is validated for an experimental dataset and compared with other image decomposition methods.

KW - Bilateral Filters

KW - component

KW - Denoising

KW - Image Classification

KW - Image Processing

KW - ISAR

KW - Target Classification

UR - http://www.scopus.com/inward/record.url?scp=85030216649&partnerID=8YFLogxK

U2 - 10.1109/RAST.2017.8003014

DO - 10.1109/RAST.2017.8003014

M3 - Conference contribution

AN - SCOPUS:85030216649

T3 - Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017

SP - 161

EP - 165

BT - Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017

A2 - Unal, M.F.

A2 - Hacioglu, A.

A2 - Yildiz, M.S.

A2 - Altan, O.

A2 - Yorukoglu, M.

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th International Conference on Recent Advances in Space Technologies, RAST 2017

Y2 - 19 June 2017 through 22 June 2017

ER -

Bozkurt H, Erer I. Information preserving preprocessing for improved radar target classification accuracy. In Unal MF, Hacioglu A, Yildiz MS, Altan O, Yorukoglu M, editors, Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 161-165. 8003014. (Proceedings of 8th International Conference on Recent Advances in Space Technologies, RAST 2017). doi: 10.1109/RAST.2017.8003014

Information preserving preprocessing for improved radar target classification accuracy

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this