An Azimuth-Continuously Controllable SAR Image Generation Algorithm Based on GAN

שמור ב:
מידע ביבליוגרפי
הוצא לאור ב:Remote Sensing vol. 17, no. 22 (2025), p. 3763-3782
מחבר ראשי: Cui Yongjie
מחברים אחרים: Liu Zhiqu, Ruan Linian, Bowen, Sheng, Wang, Ning, Xiao Xiulai, Bian Xiaolin
יצא לאור:
MDPI AG
נושאים:
גישה מקוונת:Citation/Abstract
Full Text + Graphics
Full Text - PDF
תגים: הוספת תג
אין תגיות, היה/י הראשונ/ה לתייג את הרשומה!

MARC

LEADER 00000nab a2200000uu 4500
001 3275550445
003 UK-CbPIL
022 |a 2072-4292 
024 7 |a 10.3390/rs17223763  |2 doi 
035 |a 3275550445 
045 2 |b d20250101  |b d20251231 
084 |a 231556  |2 nlm 
100 1 |a Cui Yongjie  |u Laboratory for Microwave Spatial Intelligence and Cloud Platform, Deqing Academy of Satellite Applications, Huzhou 313200, China; cyj@dasa.net.cn (Y.C.); rln@dasa.net.cn (L.R.); xxl@dasa.net.cn (X.X.); bianxl@radi.ac.cn (X.B.) 
245 1 |a An Azimuth-Continuously Controllable SAR Image Generation Algorithm Based on GAN 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a <sec sec-type="highlights"> What are the main findings? <list list-type="bullet"> <list-item> An enhanced GAN for SAR image generation, called Azimuth-Continuously Controllable Generative Adversarial Network (ACC-GAN), is proposed to enable precise interpolation between arbitrary azimuth angles. </list-item> <list-item> ACC-GAN improves the flexibility on angular generation, while maintaining the physical fidelity and angular accuracy of SAR images. </list-item> What are the implications of the main findings? <list list-type="bullet"> <list-item> Since the multi-view SAR images are very scarce and the azimuth characteristics are particularly important for SAR target recognition, the proposed ACC-GAN can provide flexible and accessible augmentation of multi-view SAR images. </list-item> The performance of deep learning models largely depends on the scale and quality of training data. However, acquiring sufficient, high-quality samples for specific observation scenarios is often challenging due to high acquisition costs. Unlike optical imagery, synthetic aperture radar (SAR) target images exhibit strong nonlinear scattering variations with changing azimuth angles, making conventional data augmentation methods such as cropping or rotation ineffective. To tackle these challenges, this paper introduces an Azimuth-Continuously Controllable Generative Adversarial Network (ACC-GAN), which incorporates a continuous azimuth conditional variable to achieve precise azimuth-controllable target generation from dual-input SAR images. Our key contributions are threefold: (1) a continuous azimuth control mechanism that enables precise interpolation between arbitrary azimuth angles; (2) a dual-discriminator framework combining similarity and azimuth supervision to ensure both visual realism and angular accuracy; and (3) conditional batch normalization integrated with adaptive feature fusion to maintain scattering consistency. Experiments on the MSTAR dataset demonstrate that ACC-GAN effectively captures nonlinear azimuth-dependent transformations, generating high-quality images that improve downstream classification accuracy and validate its practical value for SAR data augmentation. 
651 4 |a United States--US 
653 |a Deep learning 
653 |a Datasets 
653 |a Data acquisition 
653 |a Training 
653 |a Interpolation 
653 |a Generative adversarial networks 
653 |a Laboratories 
653 |a Radar imaging 
653 |a Machine learning 
653 |a Image processing 
653 |a Accuracy 
653 |a Visual discrimination 
653 |a Data augmentation 
653 |a Scattering 
653 |a Continuity (mathematics) 
653 |a Synthetic aperture radar 
653 |a Target recognition 
653 |a Neural networks 
653 |a Flexibility 
653 |a Controllability 
653 |a Image acquisition 
653 |a Algorithms 
653 |a Image quality 
653 |a Azimuth 
700 1 |a Liu Zhiqu  |u Laboratory for Microwave Spatial Intelligence and Cloud Platform, Deqing Academy of Satellite Applications, Huzhou 313200, China; cyj@dasa.net.cn (Y.C.); rln@dasa.net.cn (L.R.); xxl@dasa.net.cn (X.X.); bianxl@radi.ac.cn (X.B.) 
700 1 |a Ruan Linian  |u Laboratory for Microwave Spatial Intelligence and Cloud Platform, Deqing Academy of Satellite Applications, Huzhou 313200, China; cyj@dasa.net.cn (Y.C.); rln@dasa.net.cn (L.R.); xxl@dasa.net.cn (X.X.); bianxl@radi.ac.cn (X.B.) 
700 1 |a Bowen, Sheng  |u Laboratory of Pinghu, Jiaxing 314200, China; shengbowen@sjtu.edu.cn (B.S.); wangningtop@163.com (N.W.) 
700 1 |a Wang, Ning  |u Laboratory of Pinghu, Jiaxing 314200, China; shengbowen@sjtu.edu.cn (B.S.); wangningtop@163.com (N.W.) 
700 1 |a Xiao Xiulai  |u Laboratory for Microwave Spatial Intelligence and Cloud Platform, Deqing Academy of Satellite Applications, Huzhou 313200, China; cyj@dasa.net.cn (Y.C.); rln@dasa.net.cn (L.R.); xxl@dasa.net.cn (X.X.); bianxl@radi.ac.cn (X.B.) 
700 1 |a Bian Xiaolin  |u Laboratory for Microwave Spatial Intelligence and Cloud Platform, Deqing Academy of Satellite Applications, Huzhou 313200, China; cyj@dasa.net.cn (Y.C.); rln@dasa.net.cn (L.R.); xxl@dasa.net.cn (X.X.); bianxl@radi.ac.cn (X.B.) 
773 0 |t Remote Sensing  |g vol. 17, no. 22 (2025), p. 3763-3782 
786 0 |d ProQuest  |t Advanced Technologies & Aerospace Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3275550445/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3275550445/fulltextwithgraphics/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3275550445/fulltextPDF/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch