CIST: An Improved ISAR Imaging Method Using Convolution Neural Network

Wei, Shunjun; Liang, Jiadian; Wang, Mou; Zeng, Xiangfeng; Shi, Jun

doi:10.3390/rs12162641

Cited by 23 publications

(7 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the OMP [17], fast marginalized sparse Bayesian learning (FMSBL) [29,30], and SBRIM [21] algorithms are used as the comparison algorithms to evaluate the performance of the FBCS-RVM algorithm better. The normalized mean square error (NMSE) [31,32] measures the scattering coefficients' estimation accuracy; the smaller NMSE indicates that the estimation results of the scattering coefficients are more accurate. The target background contrast (TBR) [33] and image entropy (ENT) [34] are used to quantitatively evaluate the imaging quality.…”

Section: Results On Simulation and Experimental Datamentioning

confidence: 99%

Fast Bayesian Compressed Sensing Algorithm via Relevance Vector Machine for LASAR 3D Imaging

Tian

et al. 2021

Remote Sensing

Self Cite

View full text Add to dashboard Cite

Because of the three-dimensional (3D) imaging scene’s sparsity, compressed sensing (CS) algorithms can be used for linear array synthetic aperture radar (LASAR) 3D sparse imaging. CS algorithms usually achieve high-quality sparse imaging at the expense of computational efficiency. To solve this problem, a fast Bayesian compressed sensing algorithm via relevance vector machine (FBCS–RVM) is proposed in this paper. The proposed method calculates the maximum marginal likelihood function under the framework of the RVM to obtain the optimal hyper-parameters; the scattering units corresponding to the non-zero optimal hyper-parameters are extracted as the target-areas in the imaging scene. Then, based on the target-areas, we simplify the measurement matrix and conduct sparse imaging. In addition, under low signal to noise ratio (SNR), low sampling rate, or high sparsity, the target-areas cannot always be extracted accurately, which probably contain several elements whose scattering coefficients are too small and closer to 0 compared to other elements. Those elements probably make the diagonal matrix singular and irreversible; the scattering coefficients cannot be estimated correctly. To solve this problem, the inverse matrix of the singular matrix is replaced with the generalized inverse matrix obtained by the truncated singular value decomposition (TSVD) algorithm to estimate the scattering coefficients correctly. Based on the rank of the singular matrix, those elements with small scattering coefficients are extracted and eliminated to obtain more accurate target-areas. Both simulation and experimental results show that the proposed method can improve the computational efficiency and imaging quality of LASAR 3D imaging compared with the state-of-the-art CS-based methods.

show abstract

Section: Results On Simulation and Experimental Datamentioning

confidence: 99%

Fast Bayesian Compressed Sensing Algorithm via Relevance Vector Machine for LASAR 3D Imaging

Tian

et al. 2021

Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…According to (19), the gradient error of all convolution layers can be deduced. By combining the relationship between X and W in Equation ( 14), the gradient error of the weight term of the -th k convolution layer can be expressed as:…”

Section:  mentioning

confidence: 99%

“…Gao et al [18] proposed a residual network, which achieved the end-to-end mapping from low-resolution ISAR image to high-resolution ISAR image. The ISAR imaging method based on deep learning time-frequency analysis was proposed in [19] to achieve ISAR super-resolution imaging by improving time-frequency aggregation. Wei et al [20] proposed a new method to improve the robustness of ISAR imaging by combining the advantages of convolutional neural network (CNN) and iterative shrinkage-thresholding algorithm (ISTA).…”

Section: Introductionmentioning

confidence: 99%

ISAR High-Resolution Imaging Method With Joint FISTA and VGGNet

Xu¹,

Yang²,

Lv³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

With regard to inverse synthetic aperture radar (ISAR) imaging, traditional range-Doppler (RD) algorithm is inapplicable to sparse aperture. Although compressive sensing (CS) algorithm can overcome this problem, the imaging resolution is not high enough. When deep learning (DL) is applied to ISAR imaging, some problems may also occur, such as network complexity, the selection of labels in network training, the influence of noise and sparse aperture, and the loss of weak scattering points in network test. In order to solve the above problems, a joint fast iterative shrinkage-thresholding algorithm (FISTA) and Visual Geometry Group Network (VGGNet) high-resolution imaging method is proposed in this paper. In the proposed method, FISTA is presented to reduce the impact of noise and sparse aperture. The high-resolution processing network (HRPN) is built based on VGGNet. Then, combined with peak extraction technology, the random ideal scattering points are utilized to construct the training/validation set. Meanwhile, the training process of HRPN is analyzed theoretically, and the network test strategy is designed by the differences between the test set and the training/validation set. Extensive experiments based on both simulated and measured data demonstrate that the proposed method has good imaging performance and small network complexity. INDEX TERMSInverse synthetic aperture radar (ISAR), Deep learning (DL), Compressive sensing (CS), High-resolution processing network (HRPN), Peak extraction technology.

show abstract

“…Finally, they output the focused image of an unknown target from the trained network. Typical networks with modeldriven methods include AF-AMPNet [15], 2D-ADMM-Net (2D-ADN) [16], and convolution iterative shrinkage-thresholding (CIST) [17], etc. Although model-driven methods have strong interpretability and satisfying reconstruction performance, the optimal parameters are sensitive to SNR.…”

Section: Introductionmentioning

confidence: 99%

High-Resolution ISAR Imaging Based on Plug-and-Play 2D ADMM-Net

Bai

Zhou

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

We propose a deep learning architecture, dubbed Plug-and-play 2D ADMM-Net (PAN), by combining model-driven deep networks and data-driven deep networks for effective high-resolution 2D inverse synthetic aperture radar (ISAR) imaging with various signal-to-noise ratios (SNR) and incomplete data scenarios. First, a sparse observation model of 2D ISAR imaging is established, and a 2D ADMM algorithm is presented. On this basis, using the plug and play (PnP) technique, PnP 2D ADMM is proposed, by combining the 2D ADMM algorithm and the deep denoising network DnCNN. Then, we unroll and generalize the PnP 2D ADMM to the PAN architecture, in which all adjustable parameters in the reconstruction layers, denoising layers, and multiplier update layers are learned by end-to-end training through back-propagation. Experimental results showed that the PAN with a single parameter set can achieve noise-robust ISAR imaging with superior reconstruction performance on incomplete simulated and measured data under different SNRs.

show abstract

CIST: An Improved ISAR Imaging Method Using Convolution Neural Network

Cited by 23 publications

References 37 publications

Fast Bayesian Compressed Sensing Algorithm via Relevance Vector Machine for LASAR 3D Imaging

Fast Bayesian Compressed Sensing Algorithm via Relevance Vector Machine for LASAR 3D Imaging

ISAR High-Resolution Imaging Method With Joint FISTA and VGGNet

High-Resolution ISAR Imaging Based on Plug-and-Play 2D ADMM-Net

Contact Info

Product

Resources

About