Radar HRRP recognition based on discriminant deep autoencoders with small training data size

Pan, Mian; Jiang, Jie; Zhu, Li; Zhou, Tao

doi:10.1049/el.2016.3060

Cited by 30 publications

(13 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Radar automatic target recognition (RATR) is a specific application of the pattern recognition theory in the radar field. RATR extracts the discernible features from the frequency, phase, amplitude, and polarization information of the scattered field echo signals generated by the target in the far field of the radar, and identifies the target using the a priori target information [117][118][119][120][121]. A typical RATR system is divided into four modules: data acquisition, preprocessing, feature extraction, and selection (also known as classifier training or decision-making).…”

Section: Automatic Target Recognition Technologymentioning

confidence: 99%

“…Feature extraction isolates a set of features representing the essential attributes of the target from HRRP. Feature classification maps the feature set to the corresponding class of the target by machine learning [117][118][119][120][121][122][123][124][125][126][127][128][129][130].…”

Section: Hierarchical Target Recognition Algorithmmentioning

confidence: 99%

“…For adaptive feature extraction, we can exploit the end-to-end capability of deep neural networks. Deep learning avoids the high dependence of the traditional machine learning algorithm on human experience [118][119][120][121]. Deep neural networks include deep autoencoders, deep convolution neural networks, and deep recurrent neural networks.…”

Section: Hierarchical Target Recognition Algorithmmentioning

confidence: 99%

See 2 more Smart Citations

Advanced technology of high-resolution radar: target detection, tracking, imaging, and recognition

Liang

Liu

2019

Sci. China Inf. Sci.

View full text Add to dashboard Cite

In recent years, the performances of radar resolution, coverage, and detection accuracy have been significantly improved through the use of ultra-wideband, synthetic aperture and digital signal processing technologies. High-resolution radars (HRRs) utilize wideband signals and synthetic apertures to enhance the range and angular resolutions of tracking, respectively. They also generate one-, two-, and even threedimensional high-resolution images containing the feature information of targets, from which the targets can be precisely classified and identified. Advanced signal processing algorithms in HRRs obtain important information such as range-Doppler imaging, phase-derived ranging, and micro-motion features. However, the advantages and applications of HRRs are restricted by factors such as the reduced signal-to-noise ratio (SNR) of multi-scatter point targets, decreased tracking accuracy of multi-scatter point targets, high demands of motion compensation, and low sensitivity of the target attitude. Focusing on these problems, this paper systematically introduces the novel technologies of HRRs and discusses the issues and solutions relevant to detection, tracking, imaging, and recognition. Finally, it reviews the latest progress and representative results of HRR-based research, and suggests the future development of HRRs.

show abstract

Section: Automatic Target Recognition Technologymentioning

confidence: 99%

Section: Hierarchical Target Recognition Algorithmmentioning

confidence: 99%

Section: Hierarchical Target Recognition Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Advanced technology of high-resolution radar: target detection, tracking, imaging, and recognition

Liang

Liu

2019

Sci. China Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…By this method, problem of target recognition with unbalanced data sets is solved. In [17], a discriminative deep auto-encoder (DDAE) is proposed to improve the recognition accuracy with limited data samples.…”

Section: Introductionmentioning

confidence: 99%

Radar HRRP Target Recognition Based on Deep One-Dimensional Residual-Inception Network

Guo¹,

He²,

Wang³

et al. 2019

IEEE Access

View full text Add to dashboard Cite

A novel radar target recognition method based on the deep one-dimensional residual-inception network is proposed for a high-resolution range profile (HRRP). The traditional methods based on shallow models can hardly extract the complete information of the targets HRRP from different angles. The deep models, such as sparse autoencoder, have been adopted to solve this problem. However, these deep models with a huge amount of parameters require more training samples to guarantee the generalization performance. To solve the above-mentioned problem, a model based on the one-dimensional convolutional kernel and a pooling layer is proposed. It is worth noting that the one-dimensional convolutional kernel and pooling operation have the potential to overcome the translation sensitivity and target aspect sensitivity of the HRRP, and both of them can greatly reduce the parameters and improve the generalization performance of the model. In addition, a new loss function is proposed to further enhance the separability of features. The experimental results show that compared with other four deep models, the proposed model can achieve a good performance in recognition accuracy and robustness. INDEX TERMS Radar target recognition, high-resolution range profile, one-dimensional residual-inception network, cosine center loss.

show abstract

“…However, it is strongly supervised, this means there will be serious overfitting in small dataset. In [15], an automatic depth encoder based on discriminat is proposed to address the difficulty of small dataset. Although the method is state-of-the-art, better network structure need to be designed and the distribution characteristics of the hidden layer could still be investigated.…”

Section: Introductionmentioning

confidence: 99%

Radar HRRP Target Recognition via Semi-Supervised Multi-Task Deep Network

Zhao

Liang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Feature representation based on the high resolution range profile (HRRP) is the key technology in radar automatic target recognition(RATR). In this paper, we design a deep-u-blind denoising network(DUBDNet) to extract features with high-noise-stability. The fully convolutional DUBDNet is based on autoencoder and employs fusion layers to transfer input features to high dimensional space. Then radar HRRP shift-robust convolutional neural network(RSRNet) is proposed as the classifier. In the experiment, two radar sensors are used to measure HRRP signals of warplanes and civil airplanes. RSRNet performs high robustness to HRRP time-shift sensitivity via testing translation data. This is also a proof that convolutional neural network(CNN) is shift-robust on HRRP target recognition. Trained with noise-tonoise, DUBDNet can achieve blind-denoising in low signal-to-noise ratio(SNR) and significantly improve the correct recognition rate of targets. When the SNR of input HRRP signals is less than 5 dB, DUBDNet can increase the SNR by 10 dB. When the input SNR is −15 dB, output SNR can be increased by 15 dB and the correct recognition rate of targets can be increased by 15%.

show abstract

Radar HRRP recognition based on discriminant deep autoencoders with small training data size

Cited by 30 publications

References 11 publications

Advanced technology of high-resolution radar: target detection, tracking, imaging, and recognition

Advanced technology of high-resolution radar: target detection, tracking, imaging, and recognition

Radar HRRP Target Recognition Based on Deep One-Dimensional Residual-Inception Network

Radar HRRP Target Recognition via Semi-Supervised Multi-Task Deep Network

Contact Info

Product

Resources

About