Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification

Shouhei KIDERA; Tetsuo KIRIMOTO

doi:10.1587/transcom.E95.B.3563

Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification

Shouhei KIDERA, Tetsuo KIRIMOTO

Full Text Views

0

Cite this

Summary :

Microwave imaging techniques, in particular synthetic aperture radar (SAR), are able to obtain useful images even in adverse weather or darkness, which makes them suitable for target position or feature estimation. However, typical SAR imagery is not informative for the operator, because it is synthesized using complex radio signals with greater than 1.0 m wavelength. To deal with the target identification issue for imaging radar, various automatic target recognition (ATR) techniques have been developed. One of the most promising ATR approaches is based on neural network classification. However, in the case of SAR images heavily contaminated by random or speckle noises, the classification accuracy is severely degraded because it only compares the outputs of neurons in the final layer. To overcome this problem, this paper proposes a self organized map (SOM) based ATR method, where the binary SAR image is classified using the unified distance matrix (U-matrix) metric given by the SOM. Our numerical analyses and experiments on 5 types of civilian airplanes, demonstrate that the proposed method remarkably enhances the classification accuracy, particular in lower S/N situations, and holds a significant robustness to the angular variations of the observation.

Publication: IEICE TRANSACTIONS on Communications Vol.E95-B No.11 pp.3563-3571

Publication Date: 2012/11/01

Publicized

Online ISSN: 1745-1345

DOI: 10.1587/transcom.E95.B.3563

Type of Manuscript: PAPER

Category: Sensing

Cite this

Copy

Shouhei KIDERA, Tetsuo KIRIMOTO, "Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification" in IEICE TRANSACTIONS on Communications, vol. E95-B, no. 11, pp. 3563-3571, November 2012, doi: 10.1587/transcom.E95.B.3563.
Abstract: Microwave imaging techniques, in particular synthetic aperture radar (SAR), are able to obtain useful images even in adverse weather or darkness, which makes them suitable for target position or feature estimation. However, typical SAR imagery is not informative for the operator, because it is synthesized using complex radio signals with greater than 1.0 m wavelength. To deal with the target identification issue for imaging radar, various automatic target recognition (ATR) techniques have been developed. One of the most promising ATR approaches is based on neural network classification. However, in the case of SAR images heavily contaminated by random or speckle noises, the classification accuracy is severely degraded because it only compares the outputs of neurons in the final layer. To overcome this problem, this paper proposes a self organized map (SOM) based ATR method, where the binary SAR image is classified using the unified distance matrix (U-matrix) metric given by the SOM. Our numerical analyses and experiments on 5 types of civilian airplanes, demonstrate that the proposed method remarkably enhances the classification accuracy, particular in lower S/N situations, and holds a significant robustness to the angular variations of the observation.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E95.B.3563/_p

Copy

@ARTICLE{e95-b_11_3563,
author={Shouhei KIDERA, Tetsuo KIRIMOTO, },
journal={IEICE TRANSACTIONS on Communications},
title={Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification},
year={2012},
volume={E95-B},
number={11},
pages={3563-3571},
abstract={Microwave imaging techniques, in particular synthetic aperture radar (SAR), are able to obtain useful images even in adverse weather or darkness, which makes them suitable for target position or feature estimation. However, typical SAR imagery is not informative for the operator, because it is synthesized using complex radio signals with greater than 1.0 m wavelength. To deal with the target identification issue for imaging radar, various automatic target recognition (ATR) techniques have been developed. One of the most promising ATR approaches is based on neural network classification. However, in the case of SAR images heavily contaminated by random or speckle noises, the classification accuracy is severely degraded because it only compares the outputs of neurons in the final layer. To overcome this problem, this paper proposes a self organized map (SOM) based ATR method, where the binary SAR image is classified using the unified distance matrix (U-matrix) metric given by the SOM. Our numerical analyses and experiments on 5 types of civilian airplanes, demonstrate that the proposed method remarkably enhances the classification accuracy, particular in lower S/N situations, and holds a significant robustness to the angular variations of the observation.},
keywords={},
doi={10.1587/transcom.E95.B.3563},
ISSN={1745-1345},
month={November},}

Copy

TY - JOUR
TI - Accurate and Robust Automatic Target Recognition Method for SAR Imagery with SOM-Based Classification
T2 - IEICE TRANSACTIONS on Communications
SP - 3563
EP - 3571
AU - Shouhei KIDERA
AU - Tetsuo KIRIMOTO
PY - 2012
DO - 10.1587/transcom.E95.B.3563
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E95-B
IS - 11
JA - IEICE TRANSACTIONS on Communications
Y1 - November 2012
AB - Microwave imaging techniques, in particular synthetic aperture radar (SAR), are able to obtain useful images even in adverse weather or darkness, which makes them suitable for target position or feature estimation. However, typical SAR imagery is not informative for the operator, because it is synthesized using complex radio signals with greater than 1.0 m wavelength. To deal with the target identification issue for imaging radar, various automatic target recognition (ATR) techniques have been developed. One of the most promising ATR approaches is based on neural network classification. However, in the case of SAR images heavily contaminated by random or speckle noises, the classification accuracy is severely degraded because it only compares the outputs of neurons in the final layer. To overcome this problem, this paper proposes a self organized map (SOM) based ATR method, where the binary SAR image is classified using the unified distance matrix (U-matrix) metric given by the SOM. Our numerical analyses and experiments on 5 types of civilian airplanes, demonstrate that the proposed method remarkably enhances the classification accuracy, particular in lower S/N situations, and holds a significant robustness to the angular variations of the observation.
ER -