# ISAR Resolution Enhancement Method Exploiting Generative Adversarial Network

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. ISAR Imaging Model

## 3. Proposed GAN-Based Resolution Enhancement Method

#### 3.1. Framework of the Proposed GAN

#### 3.2. Design of the Proposed GAN

#### 3.3. Loss Function

## 4. Processing Details

#### 4.1. Data Acquisition

#### 4.2. Testing Strategy

## 5. Experiment Results and Analysis

#### 5.1. Comparison of No Noise and Full Aperture

#### 5.2. Comparison of Different SNRs

#### 5.3. Comparison of Sparse Aperture

#### 5.4. Universality and Generalization of the Proposed GAN

## 6. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Kang, L.; Luo, Y.; Zhang, Q.; Liu, X.W.; Liang, B.S. 3-D Scattering Image Sparse Reconstruction via Radar Network. IEEE Trans. Geosci. Remote Sens.
**2022**, 60, 5100414. [Google Scholar] [CrossRef] - Xue, R.H.; Bai, X.R.; Zhou, F. SAISAR-Net: A Robust Sequential Adjustment ISAR Image Classification Network. IEEE Trans. Geosci. Remote Sens.
**2021**, 60, 5214715. [Google Scholar] [CrossRef] - Bai, X.R.; Zhang, Y.; Zhou, F. High-resolution radar imaging in complex environments based on Bayesian learning with mixture models. IEEE Trans. Geosci. Remote Sens.
**2019**, 57, 972–984. [Google Scholar] [CrossRef] - Deng, L.K.; Zhang, S.H.; Zhang, C.; Liu, Y.X. A multiple-input multiple-output inverse synthetic aperture radar imaging method based on multidimensional alternating direction method of multipliers. J. Radars.
**2021**, 10, 416–431. [Google Scholar] - Liu, X.W.; Dong, C.; Luo, Y.; Kang, L.; Liu, Y.; Zhang, Q. Sparse Reconstruction for Radar Imaging Based on Quantum Algorithms. IEEE Geosci. Remote Sens. Lett.
**2022**, 19, 3507905. [Google Scholar] [CrossRef] - Cai, T.T.; Wang, L. Orthogonal Matching Pursuit for Sparse Signal Recovery with Noise. IEEE Trans. Inf. Theory
**2011**, 57, 4680–4688. [Google Scholar] [CrossRef] - Mohimani, H.; Babaie-Zadeh, M.; Jutten, C. A Fast Approach for Overcomplete Sparse Decomposition Based on Smoothed ${l}^{0}$ Norm. IEEE Trans. Signal Process.
**2009**, 57, 289–301. [Google Scholar] [CrossRef] [Green Version] - Li, S.; Amin, M.; Zhao, G.; Sun, H. Radar imaging by sparse optimization incorporating MRF clustering prior. IEEE Geosci. Remote Sens. Lett.
**2020**, 17, 1139–1143. [Google Scholar] [CrossRef] [Green Version] - Zhang, S.H.; Liu, Y.X.; Li, X. Computationally Efficient Sparse Aperture ISAR Autofocusing and Imaging Based on Fast ADMM. IEEE Trans. Geosci. Remote Sens.
**2020**, 58, 8751–8765. [Google Scholar] [CrossRef] - Gao, J.K.; Deng, B.; Qin, Y.L.; Wang, H.Q.; Li, X. Enhanced Radar Imaging Using a Complex-Valued Convolutional Neural Network. IEEE Geosci. Remote Sens. Lett.
**2019**, 16, 35–39. [Google Scholar] [CrossRef] [Green Version] - Hu, C.Y.; Wang, L.; Li, Z.; Zhu, D.Y. Inverse Synthetic Aperture Radar Imaging Using a Fully Convolutional Neural Network. IEEE Geosci. Remote Sens. Lett.
**2020**, 17, 1203–1207. [Google Scholar] [CrossRef] - Gao, X.Z.; Qin, D.; Gao, J.K. Resolution enhancement for inverse synthetic aperture radar images using a deep residual network. Microw. Opt. Technol. Lett.
**2020**, 62, 1588–1593. [Google Scholar] [CrossRef] - Yang, T.; Shi, H.Y.; Lang, M.Y.; Guo, J.W. ISAR imaging enhancement: Exploiting deep convolutional neural network for signal reconstruction. Int. J. Remote Sens.
**2020**, 41, 9447–9468. [Google Scholar] [CrossRef] - Qian, J.; Huang, S.Y.; Wang, L.; Bi, G.A.; Yang, X.B. Super-Resolution ISAR Imaging for Maneuvering Target Based on Deep-Learning-Assisted Time-Frequency Analysis. IEEE Trans. Geosci. Remote Sens.
**2022**, 60, 5201514. [Google Scholar] [CrossRef] - Wei, X.; Yang, J.; Lv, M.J.; Chen, W.F.; Ma, X.Y.; Long, M.; Xia, S.Q. ISAR High-Resolution Imaging Method With Joint FISTA and VGGNet. IEEE Access.
**2021**, 9, 86685–86697. [Google Scholar] [CrossRef] - Qin, D.; Gao, X.Z. Enhancing ISAR Resolution by a Generative Adversarial Network. IEEE Geosci. Remote Sens. Lett.
**2021**, 18, 127–131. [Google Scholar] [CrossRef] - Li, R.Z.; Zhang, S.H.; Zhang, C.; Liu, Y.X.; Li, X. Deep Learning Approach for Sparse Aperture ISAR Imaging and Autofocusing Based on Complex-Valued ADMM-Net. IEEE Sens. J.
**2021**, 21, 3437–3451. [Google Scholar] [CrossRef] - Wei, S.J.; Liang, J.D.; Wang, M.; Shi, J.; Zhang, X.L.; Ran, J.H. AF-AMPNet: A Deep Learning Approach for Sparse Aperture ISAR Imaging and Autofocusing. IEEE Trans. Geosci. Remote Sens.
**2022**, 60, 5206514. [Google Scholar] [CrossRef] - Li, X.Y.; Bai, X.R.; Zhou, F. High-Resolution ISAR Imaging and Autofocusing via 2D-ADMM-Net. Remote Sens.
**2021**, 13, 2326. [Google Scholar] [CrossRef] - Wang, X.T.; Yu, K.; Wu, S.X.; Gu, J.J.; Liu, Y.H.; Dong, C.; Qiao, Y.; Loy, C.C. ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks. In Proceedings of the European Conference on Computer Vision Workshops, Munich, Germany, 8–14 September 2018. [Google Scholar]
- Özdemir, C. Inverse Synthetic Aperture Radar Imaging with Matlab Algorithms; Wiley: Hoboken, NJ, USA, 2012. [Google Scholar]
- Xing, X.R.; Zhang, D.W. Image-to-Image Translation using a Relativistic Generative Adversarial Network. In Proceedings of the 11th International Conference on Digital Image Processing, Guangzhou, China, 10–13 May 2019. [Google Scholar] [CrossRef]
- Lim, B.; Son, S.; Kim, H.; Nah, S.; Lee, K.M. Enhanced deep residual networks for single image super-resolution. In Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Honolulu, HI, USA, 21–26 July 2017; pp. 1132–1140. [Google Scholar] [CrossRef] [Green Version]
- Ledig, C.; Theis, L.; Huszar, F.; Caballero, J.; Cunningham, A.; Acosta, A.; Aitken, A.; Tejani, A.; Totz, J.; Wang, Z.H.; et al. Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. In Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; pp. 105–114. [Google Scholar] [CrossRef] [Green Version]
- Zeng, C.Z.; Zhu, W.G.; Jia, X.; Yang, L. Sparse Aperture ISAR Imaging Method Based on Joint Constraints of Sparsity and Low Rank. IEEE Trans. Geosci. Remote Sens.
**2021**, 59, 168–181. [Google Scholar] [CrossRef]

**Figure 3.**The architecture of the proposed GAN for ISAR resolution enhancement: (

**a**) the structure diagram of generator network; (

**b**) the structure diagram of discriminator network.

**Figure 5.**Imaging results of simulated aircraft under the condition of no noise and full aperture: (

**a**) Ground truth; (

**b**) IFFT; (

**c**) GAN1; (

**d**) GAN2; (

**e**) GAN3; (

**f**) Proposed GAN.

**Figure 6.**Imaging results of Yak42 under the condition of no noise and full aperture: (

**a**) IFFT; (

**b**) GAN1; (

**c**) GAN2; (

**d**) GAN3; (

**e**) Proposed GAN.

**Figure 7.**Imaging results of simulated aircraft at SNR = 2 dB: (

**a**) Ground truth; (

**b**) IFFT; (

**c**) GAN1; (

**d**) GAN2; (

**e**) GAN3; (

**f**) Proposed GAN.

**Figure 8.**Imaging results of simulated aircraft at SNR = −4 dB: (

**a**) Ground truth; (

**b**) IFFT; (

**c**) GAN1; (

**d**) GAN2; (

**e**) GAN3; (

**f**) Proposed GAN.

**Figure 9.**Imaging results of Yak42 at SNR = 2 dB: (

**a**) IFFT; (

**b**) GAN1; (

**c**) GAN2; (

**d**) GAN3; (

**e**) Proposed GAN.

**Figure 10.**Imaging results of Yak42 at SNR = −4 dB: (

**a**) IFFT; (

**b**) GAN1; (

**c**) GAN2; (

**d**) GAN3; (

**e**) Proposed GAN.

**Figure 11.**Imaging results of simulated aircraft under sparse aperture: (

**a**) Ground truth; (

**b**) IFFT; (

**c**) GAN1; (

**d**) GAN2; (

**e**) GAN3; (

**f**) Proposed GAN.

**Figure 12.**Imaging results of Yak42 under sparse aperture: (

**a**) IFFT; (

**b**) GAN1; (

**c**) GAN2; (

**d**) GAN3; (

**e**) Proposed GAN.

Method | PSNR (dB) | SSIM | IE |
---|---|---|---|

IFFT | 15.9432 | 0.7001 | 4.3366 |

GAN1 | 22.2503 | 0.7172 | 3.0268 |

GAN2 | 26.7698 | 0.8458 | 2.0156 |

GAN3 | 26.7794 | 0.8320 | 2.1744 |

Proposed | 26.6372 | 0.8403 | 1.6376 |

Method | IFFT | GAN1 | GAN2 | GAN3 | Proposed |
---|---|---|---|---|---|

IE | 3.8012 | 2.4514 | 1.5298 | 1.9007 | 1.3477 |

Method | PSNR (dB) | SSIM | IE |
---|---|---|---|

IFFT | 16.8479 | 0.6560 | 4.4659 |

GAN1 | 22.3841 | 0.7267 | 2.6676 |

GAN2 | 26.1405 | 0.8434 | 1.9693 |

GAN3 | 26.5236 | 0.8471 | 2.1149 |

Proposed | 26.2321 | 0.8507 | 1.8374 |

Method | PSNR (dB) | SSIM | IE |
---|---|---|---|

IFFT | 15.0749 | 0.1856 | 5.3295 |

GAN1 | 22.9877 | 0.7218 | 2.7199 |

GAN2 | 26.6015 | 0.8394 | 2.0603 |

GAN3 | 26.8521 | 0.8432 | 2.1498 |

Proposed | 26.6683 | 0.8511 | 1.6324 |

Method | IFFT | GAN1 | GAN2 | GAN3 | Proposed |
---|---|---|---|---|---|

IE | 3.9670 | 2.3316 | 1.5434 | 1.8926 | 1.4509 |

Method | IFFT | GAN1 | GAN2 | GAN3 | Proposed |
---|---|---|---|---|---|

IE | 4.7616 | 2.3161 | 1.5685 | 1.8593 | 1.5270 |

Method | PSNR (dB) | SSIM | IE |
---|---|---|---|

IFFT | 15.7553 | 0.5048 | 4.8516 |

GAN1 | 22.1140 | 0.6706 | 3.1566 |

GAN2 | 25.6569 | 0.7975 | 2.1496 |

GAN3 | 26.1744 | 0.8258 | 2.4644 |

Proposed | 25.7515 | 0.8261 | 2.0390 |

Method | IFFT | GAN1 | GAN2 | GAN3 | Proposed |
---|---|---|---|---|---|

IE | 4.1299 | 2.5353 | 1.6230 | 2.0888 | 1.5077 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Wang, H.; Li, K.; Lu, X.; Zhang, Q.; Luo, Y.; Kang, L.
ISAR Resolution Enhancement Method Exploiting Generative Adversarial Network. *Remote Sens.* **2022**, *14*, 1291.
https://doi.org/10.3390/rs14051291

**AMA Style**

Wang H, Li K, Lu X, Zhang Q, Luo Y, Kang L.
ISAR Resolution Enhancement Method Exploiting Generative Adversarial Network. *Remote Sensing*. 2022; 14(5):1291.
https://doi.org/10.3390/rs14051291

**Chicago/Turabian Style**

Wang, Haobo, Kaiming Li, Xiaofei Lu, Qun Zhang, Ying Luo, and Le Kang.
2022. "ISAR Resolution Enhancement Method Exploiting Generative Adversarial Network" *Remote Sensing* 14, no. 5: 1291.
https://doi.org/10.3390/rs14051291