Галерея 3218369
![](/file/02050ab87549b8d717c55.gif)
Галерея 3218369
All Books Conferences Courses Journals & Magazines Standards Authors Citations
Proposed Azimuth-Controllable Generative Adversarial Network
Abstract: Sufficient synthetic aperture radar (SAR) target images are very important for the development of research works. However, available SAR target images are often limited i... View more
Sufficient synthetic aperture radar (SAR) target images are very important for the development of research works. However, available SAR target images are often limited in practice, which hinders the progress of SAR application. In this article, we propose an azimuth-controllable generative adversarial network to generate precise SAR target images with an intermediate azimuth between two given SAR images' azimuths. This network mainly contains three parts: 1) generator, 2) discriminator, and 3) predictor. Through the proposed specific network structure, the generator can extract and fuse the optimal target features from two input SAR target images to generate an SAR target image. Then, a similarity discriminator and an azimuth predictor are designed. The similarity discriminator can differentiate the generated SAR target images from the real SAR images to ensure the accuracy of the generated while the azimuth predictor measures the difference of azimuth between the generated and the desired to ensure the azimuth controllability of the generated. Therefore, the proposed network can generate precise SAR images, and their azimuths can be controlled well by the inputs of the deep network, which can generate the target images in different azimuths to solve the small sample problem to some degree and benefit the research works of SAR images. Extensive experimental results show the superiority of the proposed method in azimuth controllability and accuracy of SAR target image generation.
Date of Publication: 31 October 2022
TABLE I Original Number of MSTAR and Training and Testing Dataset
TABLE II Number of Training and Testing Dataset for the Evaluation of the Generation Ability
TABLE III Quantitative Results Under Increasing Azimuth Interval
TABLE IV Comparison of Quantitative Results With Other Generation Methods
TABLE V Entire Dataset of Training and Testing for SOC
TABLE VI Number of Training and Testing Dataset for the Evaluation of the Recognition Performance Under SOC Before the Data Augmentation
TABLE VII Number of Training Datasets Under EOC-D Before the Data Augmentation
TABLE VIII Number of Testing Images for EOC-D
TABLE IX Number of Training Datasets Under EOC-C and EOC-V Before the Data Augmentation
TABLE X Number of Testing Images for EOC-C
TABLE XI Number of Testing Images for EOC-V
TABLE XII Recognition Performance for Various Methods
J. C. Curlander and R. N. McDonough, Synthetic Aperture Radar, New York, NY, USA::Wiley, vol. 11, 1991.
D. E. Dudgeon and R. T. Lacoss, "An overview of automatic target recognition", Lincoln Lab. J. , vol. 6, no. 1, pp. 3-10, 1993.
S. Vitale, G. Ferraioli and V. Pascazio, "Complexity analysis of an edge preserving CNN SAR despeckling algorithm", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 6922-6925, 2020.
L. Wang, M. Zheng, W. Du, M. Wei and L. Li, "Super-resolution SAR image reconstruction via generative adversarial network", Proc. 12th Int. Symp. Antennas Propag. EM Theory , pp. 1-4, 2018.
A. Moreira, P. Prats-Iraola, M. Younis, G. Krieger, I. Hajnsek and K. P. Papathanassiou, "A tutorial on synthetic aperture radar" in IEEE Geosci. Remote Sens. Mag., vol. 1, no. 1, pp. 6-43, Mar. 2013.
S. Salcedo-Sanz, "Machine learning information fusion in earth observation: A comprehensive review of methods applications and data sources", Inf. Fusion , vol. 63, pp. 256-272, 2020.
D. Ma, X. Zhang, X. Tang, J. Ming and J. Shi, "A CNN-based method for SAR image despeckling", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 4272-4275, 2019.
M. Dai, C. Peng, A. K. Chan and D. Loguinov, "Bayesian wavelet shrinkage with edge detection for SAR image despeckling", IEEE Trans. Geosci. Remote Sens. , vol. 42, no. 8, pp. 1642-1648, Aug. 2004.
K. Doi, K. Sakurada, M. Onishi and A. Iwasaki, "GAN-based SAR-to-optical image translation with region information", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 2069-2072, 2020.
X. Wang, D. Zhu, G. Li, X.-P. Zhang and Y. He, "Proposal-copula-based fusion of spaceborne and airborne SAR images for ship target detection", Inf. Fusion , vol. 77, pp. 247-260, 2022.
B. Rasti and P. Ghamisi, "Remote sensing image classification using subspace sensor fusion", Inf. Fusion , vol. 64, pp. 121-130, 2020.
S. C. Kulkarni and P. P. Rege, "Pixel level fusion techniques for SAR and optical images: A review", Inf. Fusion , vol. 59, pp. 13-29, 2020.
G. Simone, A. Farina, F. C. Morabito, S. B. Serpico and L. Bruzzone, "Image fusion techniques for remote sensing applications", Inf. Fusion , vol. 3, no. 1, pp. 3-15, 2002.
T. D. Ross, S. W. Worrell, V. J. Velten, J. C. Mossing and M. L. Bryant, "Standard SAR ATR evaluation experiments using the MSTAR public release data set", Proc. SPIE , vol. 3370, pp. 566-573, 1998.
T. Balz and U. Stilla, "Hybrid GPU-based single-and double-bounce SAR simulation", IEEE Trans. Geosci. Remote Sens. , vol. 47, no. 10, pp. 3519-3529, Oct. 2009.
G. Franceschetti, M. Migliaccio, D. Riccio and G. Schirinzi, "SARAS: A synthetic aperture radar (SAR) raw signal simulator", IEEE Trans. Geosci. Remote Sens. , vol. 30, no. 1, pp. 110-123, Jan. 1992.
F. Zhang, X. Yao, H. Tang, Q. Yin, Y. Hu and B. Lei, "Multiple mode SAR raw data simulation and parallel acceleration for Gaofen-3 mission", IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. , vol. 11, no. 6, pp. 2115-2126, Jun. 2018.
M. J. Collins and J. M. Allan, "Modeling and simulation of SAR image texture", IEEE Trans. Geosci. Remote Sens. , vol. 47, no. 10, pp. 3530-3546, Oct. 2009.
M. Zhang, Z. Cui, X. Wang and Z. Cao, "Data augmentation method of SAR image dataset", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 5292-5295, 2018.
Z. Luo, X. Jiang and X. Liu, "Synthetic minority class data by generative adversarial network for imbalanced SAR target recognition", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 2459-2462, 2020.
K. El-Darymli, E. W. Gill, P. McGuire, D. Power and C. Moloney, "Automatic target recognition in synthetic aperture radar imagery: A state-of-the-art review", IEEE Access , vol. 4, pp. 6014-6058, 2016.
M. Wilmanski, C. Kreucher and J. Lauer, "Modern approaches in deep learning for SAR ATR", Proc. SPIE , vol. 9843, 2016.
M. Gong, H. Yang and P. Zhang, "Feature learning and change feature classification based on deep learning for ternary change detection in SAR images", ISPRS J. Photogrammetry Remote Sens. , vol. 129, pp. 212-225, 2017.
W. Pu, "Shuffle GAN with autoencoder: A deep learning approach to separate moving and stationary targets in SAR imagery", IEEE Trans. Neural Netw. Learn. Syst. , vol. 33, no. 9, pp. 4770-4784, Sep. 2022.
J. Guo, B. Lei, C. Ding and Y. Zhang, "Synthetic aperture radar image synthesis by using generative adversarial nets", IEEE Geosci. Remote Sens. Lett. , vol. 14, no. 7, pp. 1111-1115, Jul. 2017.
Z. Cui, M. Zhang, Z. Cao and C. Cao, "Image data augmentation for SAR sensor via generative adversarial nets", IEEE Access , vol. 7, pp. 42255-42268, 2019.
T. Jiang, Z. Cui, Z. Zhou and Z. Cao, "Data augmentation with Gabor filter in deep convolutional neural networks for SAR target recognition", Proc. IEEE Int. Geosci. Remote Sens. Symp. , pp. 689-692, 2018.
C. Zheng, X. Jiang and X. Liu, "Semi-supervised SAR ATR via multi-discriminator generative adversarial network", IEEE Sensors J. , vol. 19, no. 17, pp. 7525-7533, Sep. 2019.
S. Du, J. Hong, Y. Wang and Y. Qi, "A high-quality multicategory SAR images generation method with multiconstraint GAN for ATR", IEEE Geosci. Remote Sens. Lett. , vol. 19, Mar. 2021.
C. Mao, L. Huang, Y. Xiao, F. He and Y. Liu, "Target recognition of SAR image based on CN-GAN and CNN in complex environment", IEEE Access , vol. 9, pp. 39608-39617, 2021.
S. Saha, F. Bovolo and L. Bruzzone, "Building change detection in VHR SAR images via unsupervised deep transcoding", IEEE Trans. Geosci. Remote Sens. , vol. 59, no. 3, pp. 1917-1929, Mar. 2021.
J. Pei, Y. Huang, W. Huo, Y. Zhang, J. Yang and T.-S. Yeo, "SAR automatic target recognition based on multiview deep learning framework", IEEE Trans. Geosci. Remote Sens. , vol. 56, no. 4, pp. 2196-2210, Apr. 2018.
J. Pei, Y. Huang, W. Huo, J. Wu, J. Yang and H. Yang, "SAR imagery feature extraction using 2DPCA-based two-dimensional neighborhood virtual points discriminant embedding", IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. , vol. 9, no. 6, pp. 2206-2214, Jun. 2016.
I. Goodfellow, "Generative adversarial networks", Commun. ACM , vol. 63, no. 11, pp. 139-144, Oct. 2020.
K. He, X. Zhang, S. Ren and J. Sun, "Deep residual learning for image recognition", Proc. IEEE Conf. Comput. Vis. Pattern Recognit. , pp. 770-778, 2016.
S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift", Proc. Int. Conf. Mach. Learn. , pp. 448-456, 2015.
A. L. Maas, A. Y. Hannun and A. Y. Ng, "Rectifier nonlinearities improve neural network acoustic models", Proc. Int. Conf. Mach. Learn. , 2013.
M. Arjovsky, S. Chintala and L. Bottou, "Wasserstein generative adversarial networks", Proc. Int. Conf. Mach. Learn. , pp. 214-223, 2017.
J. Dai, "Deformable convolutional networks", Proc. IEEE Int. Conf. Comput. Vis. , pp. 764-773, 2017.
C. J. Willmott and K. Matsuura, "Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance", Climate Res. , vol. 30, no. 1, pp. 79-82, 2005.
Z. Wang, A. C. Bovik, H. R. Sheikh and E. P. Simoncelli, "Image quality assessment: From error visibility to structural similarity", IEEE Trans. Image Process. , vol. 13, no. 4, pp. 600-612, Apr. 2004.
S. Kandadai, J. Hardin and C. D. Creusere, "Audio quality assessment using the mean structural similarity measure", Proc. IEEE Int. Conf. Acoust. Speech Signal Process. , pp. 221-224, 2008.
A. Radford, L. Metz and S. Chintala, "Unsupervised representation learning with deep convolutional generative adversarial networks", 2015.
S. Chen, H. Wang, F. Xu and Y.-Q. Jin, "Target classification using the deep convolutional networks for SAR images", IEEE Trans. Geosci. Remote Sens. , vol. 54, no. 8, pp. 4806-4817, Aug. 2016.
D. A. Morgan, "Deep convolutional neural networks for ATR from SAR imagery" in Proc. SPIE, vol. 9475, 2015.
J. Ding, B. Chen, H. Liu and M. Huang, "Convolutional neural network with data augmentation for SAR target recognition", IEEE Geosci. Remote Sens. Lett. , vol. 13, no. 3, pp. 364-368, Mar. 2016.
Y. Li, X. Li, Q. Sun and Q. Dong, "SAR image classification using CNN embeddings and metric learning", IEEE Geosci. Remote Sens. Lett. , vol. 19, Sep. 2020.
I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin and A. C. Courville, "Improved training of Wasserstein GANs", Proc. Adv. Neural Inf. Process. Syst. , pp. 5769-5779, 2017.
W. Zhang, Y. Zhu and Q. Fu, "Semi-supervised deep transfer learning-based on adversarial feature learning for label limited SAR target recognition", IEEE Access , vol. 7, pp. 152412-152420, 2019.
F. Gao, Q. Liu, J. Sun, A. Hussain and H. Zhou, "Integrated GANs: Semi-supervised SAR target recognition", IEEE Access , vol. 7, pp. 113999-114013, 2019.
J. Guan, J. Liu, P. Feng and W. Wang, "Multiscale deep neural network with two-stage loss for SAR target recognition with small training set", IEEE Geosci. Remote Sens. Lett. , vol. 19, 2022.
Y. Zhai, W. Zhou, B. Sun, J. Li and F. Scotti, "Weakly contrastive learning via batch instance discrimination and feature clustering for small sample SAR ATR", IEEE Trans. Geosci. Remote Sens. , vol. 60, 2022.
IEEE Account
Change Username/Password
Update Address
Purchase Details
Payment Options
Order History
View Purchased Documents
Need Help?
US & Canada: +1 800 678 4333
Worldwide: +1 732 981 0060
Contact & Support
Synthetic aperture radar (SAR) is an important microwave remote sensing system in both mechanism and application, which has the ability to obtain high-resolution images with the pulse compression technology and synthetic aperture principle [1] , [2] . It can obtain more distinct information of the target and the scene, for instance, geometry, material, and structure, than optical sensors, infrared sensors, etc. With the high-resolution coherent imaging capability of all-weather, all day, and penetration, SAR has been widely used in the field of geography, remote sensing, military fields, etc.
For the development of the theory and technology of SAR, research works have been carried out in many fields of SAR, such as SAR image despeckling [3] , superresolution [4] , target detection, classification, recognition, and multisensor image fusion [5] , [6] . All these research works are driven by SAR data among which the SAR target images are the most important. For example, in the field of SAR image despeckling [7] , SAR images are necessary for the research works on the SAR speckle characteristics and the despeckling algorithm of both traditional methods and deep learning [8] . In the field of SAR multisensor image fusion [9] , [10] , [11] and pixel image fusion for SAR and optical images [12] all require high-quality SAR images for better fusing the scene information and the interpretation of SAR scene images [13] . As the most representative, in the field of SAR automatic target recognition (ATR), a great quantity of SAR target images is necessary for the acquirement of target features, improvement of the recognition ratio, and promotion of the practical application of SAR ATR [14] .
However, in the actual situation, an abundant number of SAR images are lacking, and SAR image acquisition is difficult and consumes resources. Even if there are some SAR images, they are likely obtained by different imaging conditions, such as the band, platform, azimuth, etc., and these SAR images cannot contain enough information of the scene or target for the research works of SAR fields. The insufficiency of SAR image data or lack of the characteristics of scene or target in SAR images has become a great obstacle of almost all SAR fields and hinders the progress of SAR applications [15] .
To solve this problem, many kinds of research are carried out in recent years. And there are mainly three types of SAR target image acquirement: 1) measured data collection, 2) electromagnetic simulation, and 3) sample augmentation [16] , [17] . First of all, measured data collection can obtain the SAR target images under different actual scenarios with different platforms. These acquired data are the most authentic and effective. However, this acquirement will consume massive resources of human, material, and time, and the number of the acquired SAR target images in each experiment is often limited. The result is that it cannot be used as a cost-effective way for the application to obtain enough SAR data.
Through the 3-D modeling of the target and electromagnetic calculation imaging, the electromagnetic simulation is, although not as accurate as the real SAR data, comparatively accurate [18] . The results of the simulation SAR target images are related to how precise the 3-D models and electromagnetic calculation methods are. But the more accurate the 3-D model and electromagnetic calculation method are, the greater the computation will be and the slower the compute process is, which can lead to the massive consumption of the time resource. Besides, when every single different radar parameter gets changed, the computation of electromagnetic simulation needs to start from scratch without using prior knowledge of existing simulation data.
As a result, although electromagnetic simulation is an alternative approach, it cannot also be an effective way to solve the lack of radar data amount. Last of all, sample augmentation, mainly employed in the field of SAR ATR, is to increase the diversity of the SAR sample and avoid overfitting of the classifier, such as translation, rotation, adding-noise, etc. [19] . Luo et al. [20] proposed a synthetic minority class data method for improving imbalanced SAR target recognition using the generative adversarial network (GAN). However, these methods are only the augmentation from the view of image processing, and many augmented images do not conform to the law of radar imaging and do not contain new information. Therefore, it cannot increase the intrinsic information of the target essentially.
In recent years, when deep learning has been applied in signal and image processing fields and demonstrated its superior performance, lots of excellent scholars mainly focus on the SAR image generation and have proposed several deep learning methods with outstanding results [21] , [22] , [23] , [24] . For example, Guo et al. [25] proposed a conditional generative adversarial net (CGAN) with a clutter normalization method to ease the model collapse during the generation of SAR images. Cui et al. [26] proposed a deep convolutional GAN (DCGAN) to generate SAR images with random azimuths and employed an azimuth discriminator to filter the desired generated images with the azimuths close to specific angles. Jiang et al. [27] proposed a Gabor-Deep Convolutional Neural Networks (G-DCNNs), which is a method of data augmentation with a Gabor filter in DCNNs for SAR ATR. It overcame the severe overfitting due to limited SAR image training data when applying DCNNs. Zheng et al. [28] proposed a multidiscriminator GAN with a label smoothing regularization to generate SAR target images with unclear types. Du et al. [29] proposed a multiconstraint GAN (MCGAN) to generate high-quality multicategory SAR images to address the poor image quality problem. Mao et al. [30] combined Constrained Naive Generative Adversarial Networks (CN-GAN) with least squares generative adversarial networks and image-to-image translation to address the problem of low signal-to-clutter-noise ratio, model instability, and the excessive freedom degree of the output, which appeared in conventional GAN. Saha et al. [31] employed a transfer learning framework using a cycle-consistent generative adversarial network (CycleGAN) to train for the suboptimal task of transcoding SAR images into optical images. These existing deep-learning SAR image generators greatly promote the research of SAR image acquirement.
However, most of the current SAR image generation methods are just the augmentation of the SAR dataset and generated from random noise, which means these methods can only generate abundant SAR images without controlling the azimuths. When the azimuth distribution of the SAR dataset is not balanced, the generated SAR image dataset by current generation methods is concentrated around certain azimuths. In practice, the features of the target in the SAR image are changing when the azimuth of the SAR image is different. And the lacking of azimuth in the SAR image dataset is actually equivalent to the lacking of the features of the target, which can negatively influence the recognition results and other research works [32] . Therefore, the azimuth-controllable generation of the SAR target images is beneficial and necessary for the improvement of target recognition and other research works. It is still a gap between the practical demands and the current methods.
Therefore, we proposed an azimuth-controllable generative adversarial network in this article, which can generate precise SAR images with an intermediate azimuth between two given
Сисястая блондинка госпожа
Галерея 2646133
Зрелая итальянка Карла часто показывает себя голой по вебке