Feature Learning for SAR Target Recognition with Unknown Classes by Using CVAE-GAN

Even though deep learning (DL) has achieved excellent results on some public data sets for synthetic aperture radar (SAR) automatic target recognition(ATR), several problems exist at present. One is the lack of transparency and interpretability for most of the existing DL networks. Another is the neglect of unknown target classes which are often present in practice. To solve the above problems, a deep generation as well as recognition model is derived based on Conditional Variational Auto-encoder (CVAE) and Generative Adversarial Network (GAN). A feature space for SAR-ATR is built based on the proposed CVAE-GAN model. By using the feature space, clear SAR images can be generated with given class labels and observation angles. Besides, the feature of the SAR image is continuous in the feature space and can represent some attributes of the target. Furthermore, it is possible to classify the known classes and reject the unknown target classes by using the feature space. Experiments on the MSTAR data set validate the advantages of the proposed method.

Download Full-text

Underwater Acoustic Target Recognition Based on Generative Adversarial Network Data Augmentation

INTER-NOISE and NOISE-CON Congress and Conference Proceedings ◽

10.3397/in-2021-2737 ◽

2021 ◽

Vol 263 (2) ◽

pp. 4558-4564

Author(s):

Minghong Zhang ◽

Xinwei Luo

Keyword(s):

Data Augmentation ◽

Target Recognition ◽

Training Data ◽

Small Samples ◽

Generative Adversarial Network ◽

Data Set ◽

Underwater Acoustic ◽

Adversarial Network ◽

Acoustic Target ◽

The Impact

Underwater acoustic target recognition is an important aspect of underwater acoustic research. In recent years, machine learning has been developed continuously, which is widely and effectively applied in underwater acoustic target recognition. In order to acquire good recognition results and reduce the problem of overfitting, Adequate data sets are essential. However, underwater acoustic samples are relatively rare, which has a certain impact on recognition accuracy. In this paper, in addition of the traditional audio data augmentation method, a new method of data augmentation using generative adversarial network is proposed, which uses generator and discriminator to learn the characteristics of underwater acoustic samples, so as to generate reliable underwater acoustic signals to expand the training data set. The expanded data set is input into the deep neural network, and the transfer learning method is applied to further reduce the impact caused by small samples by fixing part of the pre-trained parameters. The experimental results show that the recognition result of this method is better than the general underwater acoustic recognition method, and the effectiveness of this method is verified.

Download Full-text

GAINESIS: Generative Artificial Intelligence NEtlists SynthesIS

Electronics ◽

10.3390/electronics11020245 ◽

2022 ◽

Vol 11 (2) ◽

pp. 245

Author(s):

Konstantinos G. Liakos ◽

Georgios K. Georgakilas ◽

Fotis C. Plessas ◽

Paris Kitsos

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Power Analysis ◽

Public Libraries ◽

Data Sets ◽

Hardware Trojan ◽

Generative Adversarial Network ◽

Data Set ◽

Encrypted Data ◽

Adversarial Network

A significant problem in the field of hardware security consists of hardware trojan (HT) viruses. The insertion of HTs into a circuit can be applied for each phase of the circuit chain of production. HTs degrade the infected circuit, destroy it or leak encrypted data. Nowadays, efforts are being made to address HTs through machine learning (ML) techniques, mainly for the gate-level netlist (GLN) phase, but there are some restrictions. Specifically, the number and variety of normal and infected circuits that exist through the free public libraries, such as Trust-HUB, are based on the few samples of benchmarks that have been created from circuits large in size. Thus, it is difficult, based on these data, to develop robust ML-based models against HTs. In this paper, we propose a new deep learning (DL) tool named Generative Artificial Intelligence Netlists SynthesIS (GAINESIS). GAINESIS is based on the Wasserstein Conditional Generative Adversarial Network (WCGAN) algorithm and area–power analysis features from the GLN phase and synthesizes new normal and infected circuit samples for this phase. Based on our GAINESIS tool, we synthesized new data sets, different in size, and developed and compared seven ML classifiers. The results demonstrate that our new generated data sets significantly enhance the performance of ML classifiers compared with the initial data set of Trust-HUB.

Download Full-text

Automatic Target Recognition for Synthetic Aperture Radar Images Based on Super-Resolution Generative Adversarial Network and Deep Convolutional Neural Network

Remote Sensing ◽

10.3390/rs11020135 ◽

2019 ◽

Vol 11 (2) ◽

pp. 135 ◽

Cited By ~ 5

Author(s):

Xiaoran Shi ◽

Feng Zhou ◽

Shuang Yang ◽

Zijing Zhang ◽

Tao Su

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Target Recognition ◽

Super Resolution ◽

Automatic Target Recognition ◽

Synthetic Aperture ◽

Sar Image ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Feature Characterization

Aiming at the problem of the difficulty of high-resolution synthetic aperture radar (SAR) image acquisition and poor feature characterization ability of low-resolution SAR image, this paper proposes a method of an automatic target recognition method for SAR images based on a super-resolution generative adversarial network (SRGAN) and deep convolutional neural network (DCNN). First, the threshold segmentation is utilized to eliminate the SAR image background clutter and speckle noise and accurately extract target area of interest. Second, the low-resolution SAR image is enhanced through SRGAN to improve the visual resolution and the feature characterization ability of target in the SAR image. Third, the automatic classification and recognition for SAR image is realized by using DCNN with good generalization performance. Finally, the open data set, moving and stationary target acquisition and recognition, is utilized and good recognition results are obtained under standard operating condition and extended operating conditions, which verify the effectiveness, robustness, and good generalization performance of the proposed method.

Download Full-text

Pulsar candidate classification using generative adversary networks

Monthly Notices of the Royal Astronomical Society ◽

10.1093/mnras/stz2975 ◽

2019 ◽

Vol 490 (4) ◽

pp. 5424-5439 ◽

Cited By ~ 3

Author(s):

Ping Guo ◽

Fuqing Duan ◽

Pei Wang ◽

Yao Yao ◽

Qian Yin ◽

...

Keyword(s):

Class Imbalance ◽

Feature Learning ◽

Computer Experiments ◽

Support Vector ◽

Data Sets ◽

Class Imbalance Problem ◽

Generative Adversarial Network ◽

Feature Representations ◽

Adversarial Network ◽

Imbalance Problem

ABSTRACT Discovering pulsars is a significant and meaningful research topic in the field of radio astronomy. With the advent of astronomical instruments, the volume and rate of data acquisition have grown exponentially. This development necessitates a focus on artificial intelligence (AI) technologies that can mine large astronomical data sets. Automatic pulsar candidate identification (APCI) can be considered as a task determining potential candidates for further investigation and eliminating the noise of radio-frequency interference and other non-pulsar signals. As reported in the existing literature, AI techniques, especially convolutional neural network (CNN)-based techniques, have been adopted for APCI. However, it is challenging to enhance the performance of CNN-based pulsar identification because only an extremely limited number of real pulsar samples exist, which results in a crucial class imbalance problem. To address these problems, we propose a framework that combines a deep convolution generative adversarial network (DCGAN) with a support vector machine (SVM). The DCGAN is used as a sample generation and feature learning model, and the SVM is adopted as the classifier for predicting the label of a candidate at the inference stage. The proposed framework is a novel technique, which not only can solve the class imbalance problem but also can learn the discriminative feature representations of pulsar candidates instead of computing hand-crafted features in the pre-processing steps. The proposed method can enhance the accuracy of the APCI, and the computer experiments performed on two pulsar data sets verified the effectiveness and efficiency of the proposed method.

Download Full-text

Geometric Algebra Neuron for SAR Automation Target Recognition

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.187.319 ◽

2011 ◽

Vol 187 ◽

pp. 319-325

Author(s):

Wen Ming Cao ◽

Xiong Feng Li ◽

Li Juan Pu

Keyword(s):

Geometric Algebra ◽

Target Recognition ◽

Dimensional Space ◽

Feature Space ◽

Automatic Target Recognition ◽

High Dimensional ◽

Small Data ◽

Data Set ◽

High Dimensional Space ◽

Stationary Target

Biometric Pattern Recognition aim at finding the best coverage of per kind of sample’s distribution in the feature space. This paper employed geometric algebra to determine local continuum (connected) direction and connected path of same kind of target of SAR images of the complex geometrical body in high dimensional space. We researched the property of the GA Neuron of the coverage body in high dimensional space and studied a kind of SAR ATR(SAR automatic target recognition) technique which works with small data amount and result to high recognizing rate. Finally, we verified our algorithm with MSTAR (Moving and Stationary Target Acquisition and Recognition) [1] data set.

Download Full-text

Adversarial Attack for SAR Target Recognition Based on UNet-Generative Adversarial Network

Remote Sensing ◽

10.3390/rs13214358 ◽

2021 ◽

Vol 13 (21) ◽

pp. 4358

Author(s):

Chuan Du ◽

Lei Zhang

Keyword(s):

Target Recognition ◽

Automatic Target Recognition ◽

Reconstruction Error ◽

Sar Images ◽

Generative Adversarial Network ◽

Adversarial Network ◽

Scattering Centers ◽

Weak Scattering ◽

Adversarial Examples ◽

Adversarial Attack

Some recent articles have revealed that synthetic aperture radar automatic target recognition (SAR-ATR) models based on deep learning are vulnerable to the attacks of adversarial examples and cause security problems. The adversarial attack can make a deep convolutional neural network (CNN)-based SAR-ATR system output the intended wrong label predictions by adding small adversarial perturbations to the SAR images. The existing optimization-based adversarial attack methods generate adversarial examples by minimizing the mean-squared reconstruction error, causing smooth target edge and blurry weak scattering centers in SAR images. In this paper, we build a UNet-generative adversarial network (GAN) to refine the generation of the SAR-ATR models’ adversarial examples. The UNet learns the separable features of the targets and generates the adversarial examples of SAR images. The GAN makes the generated adversarial examples approximate to real SAR images (with sharp target edge and explicit weak scattering centers) and improves the generation efficiency. We carry out abundant experiments using the proposed adversarial attack algorithm to fool the SAR-ATR models based on several advanced CNNs, which are trained on the measured SAR images of the ground vehicle targets. The quantitative and qualitative results demonstrate the high-quality adversarial example generation and excellent attack effectiveness and efficiency improvement.

Download Full-text

Time-Frequency Analysis and Target Recognition of HRRP Based on CN-LSGAN, STFT, and CNN

Complexity ◽

10.1155/2021/6664530 ◽

2021 ◽

Vol 2021 ◽

pp. 1-10

Author(s):

Jianghua Nie ◽

Yongsheng Xiao ◽

Lizhen Huang ◽

Feng Lv

Keyword(s):

Least Squares ◽

Data Augmentation ◽

Short Term Memory ◽

Target Recognition ◽

Signal To Noise Ratio ◽

Recognition Performance ◽

Feature Learning ◽

Generative Adversarial Network ◽

Time Frequency ◽

Adversarial Network

Aiming at the problem of radar target recognition of High-Resolution Range Profile (HRRP) under low signal-to-noise ratio conditions, a recognition method based on the Constrained Naive Least-Squares Generative Adversarial Network (CN-LSGAN), Short-time Fourier Transform (STFT), and Convolutional Neural Network (CNN) is proposed. Combining the Least-Squares Generative Adversarial Network (LSGAN) with the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP), the CN-LSGAN is presented and applied to the HRRP denoise. The frequency domain and phase features of HRRP are gained by STFT in order to facilitate feature learning and also match the input data format of the CNN. These experimental results show that the CN-LSGAN has better data augmentation performance and can effectively avoid the model collapse compared to the generative adversarial network (GAN) and LSGAN. Also, the method has better recognition performance than the one-dimensional CNN method and the Long Short-Term Memory (LSTM) network method.

Download Full-text

Multi-Block Mixed Sample Semi-Supervised Learning for SAR Target Recognition

Remote Sensing ◽

10.3390/rs13030361 ◽

2021 ◽

Vol 13 (3) ◽

pp. 361

Author(s):

Ye Tian ◽

Jianguo Sun ◽

Pengyuan Qi ◽

Guisheng Yin ◽

Liguo Zhang

Keyword(s):

Supervised Learning ◽

Target Recognition ◽

Recognition Performance ◽

Automatic Target Recognition ◽

Training Image ◽

Data Sets ◽

Data Set ◽

Stationary Target ◽

Mixing Method ◽

Sar Target Recognition

In recent years, synthetic aperture radar (SAR) automatic target recognition has played a crucial role in multiple fields and has received widespread attention. Compared with optical image recognition with massive annotation data, lacking sufficient labeled images limits the performance of the SAR automatic target recognition (ATR) method based on deep learning. It is expensive and time-consuming to annotate the targets for SAR images, while it is difficult for unsupervised SAR target recognition to meet the actual needs. In this situation, we propose a semi-supervised sample mixing method for SAR target recognition, named multi-block mixed (MBM), which can effectively utilize the unlabeled samples. During the data preprocessing stage, a multi-block mixed method is used to interpolate a small part of the training image to generate new samples. Then, the new samples are used to improve the recognition accuracy of the model. To verify the effectiveness of the proposed method, experiments are carried out on the moving and stationary target acquisition and recognition (MSTAR) data set. The experimental results fully demonstrate that the proposed MBM semi-supervised learning method can effectively address the problem of annotation insufficiency in SAR data sets and can learn valuable information from unlabeled samples, thereby improving the recognition performance.

Download Full-text

Exploration of Campus Layout Based on Generative Adversarial Network

Proceedings of the 2020 DigitalFUTURES ◽

10.1007/978-981-33-4400-6_16 ◽

2021 ◽

pp. 169-178

Author(s):

Yubo Liu ◽

Yihua Luo ◽

Qiaoming Deng ◽

Xuanxing Zhou

Keyword(s):

Deep Learning ◽

Experimental Results ◽

Data Sets ◽

Generative Adversarial Network ◽

Data Set ◽

Effective Screening ◽

Adversarial Network ◽

Sample Data ◽

Layout Generation ◽

The Given

AbstractThis paper aims to explore the idea and method of using deep learning with a small amount sample to realize campus layout generation. From the perspective of the architect, we construct two small amount sample campus layout data sets through artificial screening with the preference of the specific architects. These data sets are used to train the ability of Pix2Pix model to automatically generate the campus layout under the condition of the given campus boundary and surrounding roads. Through the analysis of the experimental results, this paper finds that under the premise of effective screening of the collected samples, even using a small amount sample data set for deep learning can achieve a good result.

Download Full-text

Nonlinear optimization generating the Tomb Mural Blocks by GANS

Applied Mathematics and Nonlinear Sciences ◽

10.2478/amns.2020.2.00072 ◽

2020 ◽

Vol 0 (0) ◽

Author(s):

Meng Wu ◽

Adim Payshanbiev ◽

Qing Zhao ◽

Wenzong Yang

Keyword(s):

Image Inpainting ◽

Special Kind ◽

Training Data ◽

Data Sets ◽

Generative Adversarial Network ◽

Data Set ◽

Adversarial Network ◽

Defect Area ◽

Digital Restoration ◽

Nonlinear Feature

AbstractTomb murals are the special kind of murals that are buried underground. Due to the narrow exit of the tomb passage, the tomb murals were excavated by dividing the whole mural into blocks, which made lots of information missing between the blocks. The digital restoration technology Image inpainting uses the edge information around the missing parts to spread the information inside of the defect area and fills the information from the outside to the inside. But it is not suitable for filling the missing parts between the tomb mural blocks. Because these parts are large for exemplar-based inpainting which may make texture dislocation and for PDE which may make cartoon blur. It is a need to generate the information outwards to complete the information. The generative adversarial network uses deep learning training by the murals remains to generate the information from inside to outside, but the typical GAN doesn‘t have a good nonlinear feature. This paper provided a generating technology based on the deep convolution generative adversarial network to rebuild the missing information between the tomb mural blocks. It built the training data set of the simulation platform with Keras and designed a whole mural generation scheme based on DCGAN. In order to get better generated results to avoid the bad artifacts; it adds the nonlinear layers by choosing 13 layers convolution and 2 deconvolution layers of the generator and contained 5 layers convolution discriminator; it designed a new phased nonlinear loss function by using Pycharm pretreatment for Numpy array file data sets; finally, it completed the generate tomb mural information to obtain the good simulation effect.

Download Full-text