In an attempt to exploit the automatic feature extraction ability of biologically-inspired deep learning models, and enhance the learning of target features, we propose a novel deep learning algorithm. This is based on a deep convolutional neural network (DCNN) trained with an improved cost function, and combined with a support vector machine (SVM). Specifically, class separation information, which explicitly facilitates intra-class compactness and inter-class separability in the process of learning features, is added to an improved cost function as a regularization term, to enhance the DCNN’s feature extraction ability. The enhanced DCNN is applied to learn the features of Synthetic Aperture Radar (SAR) images, and the SVM is utilized to map features into output labels. Simulation experiments are performed using benchmark SAR image data from the Moving and Stationary Target Acquisition and Recognition (MSTAR) database. Comparative results demonstrate the effectiveness of our proposed method, with an average accuracy of 99% on ten types of targets, including variants and articulated targets. We conclude that our proposed DCNN method has significant potential to be exploited for SAR image target recognition, and can serve as a new benchmark for the research community.

/pdf/a-new-algorithm-for-sar-image-target-recognition-based-on-an-2vay209ssd.pdf

A New Algorithm for SAR Image Target Recognition Based on an Improved Deep Convolutional Neural Network

Among many synthetic aperture radar (SAR) automatic target recognition (ATR) algorithms, convolutional neural network (CNN)-based algorithms are the commonly used methods. However, most previous SA...

SAR ATR based on displacement- and rotation-insensitive CNN

Neural network such as an autoencoder (AE) and a convolutional autoencoder (CAE) have been successfully applied in image feature extraction. For the statistical distribution of polarimetric synthetic aperture radar (POLSAR) data, we combine the Wishart distance measurement into the training process of the AE and the CAE. In this paper, a new type of AE and CAE is specially defined, which we name them Wishart-AE (WAE) and Wishart-CAE (WCAE). Furthermore, we connect the WAE or the WCAE with a softmax classifier to compose a classification model for the purpose of POLSAR image classification. Compared with AE and CAE models, WAE and WCAE models can achieve higher classification accuracy because they could obtain the classification features, which are more suitable for POLSAR data. What is more, the WCAE model utilizes the local spatial information of a POLSAR image when compared with the WAE model. A convolutional natural network (CNN), which also makes use of the spatial information, has been widely applied in image classification, but our WCAE model is time-saving than the CNN model. Given the above, our methods not only improve the classification performance but also save the experimental time. Experimental results on four POLSAR datasets also demonstrate that our proposed methods are significantly effective.

POLSAR Image Classification via Wishart-AE Model or Wishart-CAE Model

Despite the fact that automatic target recognition (ATR) in Synthetic aperture radar (SAR) images has been extensively researched due to its practical use in both military and civil applications, it remains an unsolved problem. The major challenges of ATR in SAR stem from severe data scarcity and great variation of SAR images. Recent work started to adopt convolutional neural networks (CNNs), which, however, remain unable to handle the aforementioned challenges due to their high dependency on large quantities of data. In this paper, we propose a novel deep convolutional learning architecture, called Multi-Stream CNN (MS-CNN), for ATR in SAR by leveraging SAR images from multiple views. Specifically, we deploy a multi-input architecture that fuses information from multiple views of the same target in different aspects; therefore, the elaborated multi-view design of MS-CNN enables it to make full use of limited SAR image data to improve recognition performance. We design a Fourier feature fusion framework derived from kernel approximation based on random Fourier features which allows us to unravel the highly nonlinear relationship between images and classes. More importantly, MS-CNN is qualified with the desired characteristic of easy and quick manoeuvrability in real SAR ATR scenarios, because it only needs to acquire real-time GPS information from airborne SAR to calculate aspect differences used for constructing testing samples. The effectiveness and generalization ability of MS-CNN have been demonstrated by extensive experiments under both the Standard Operating Condition (SOC) and Extended Operating Condition (EOC) on the MSTAR dataset. Experimental results have shown that our proposed MS-CNN can achieve high recognition rates and outperform other state-of-the-art ATR methods.

Multi-Stream Convolutional Neural Network for SAR Automatic Target Recognition

The outstanding pattern recognition performance of deep learning brings new vitality to the synthetic aperture radar (SAR) automatic target recognition (ATR). However, there is a limitation in current deep learning based ATR solution that each learning process only handles one SAR image, namely learning the static scattering information, while missing the space-varying information. It is obvious that space-varying scattering information introduced in the multi-aspect joint recognition should improve the classification accuracy and robustness. In this paper, a novel multi-aspect-aware method is proposed to achieve this idea through the bidirectional long short-term memory (LSTM) recurrent neural networks-based space-varying scattering information learning. Specifically, we first select different aspect images to generate the multi-aspect space-varying image sequences. Then, the Gabor filter and three-patch local binary pattern are progressively implemented to extract comprehensive spatial features, followed by dimensionality reduction with the multi-layer perceptron network. Finally, we design a bidirectional LSTM recurrent neural network to learn the multi-aspect features with further integrating the softmax classifier to achieve target recognition. Experimental results demonstrate that the proposed method can achieve 99.9% accuracy for 10-class recognition. Besides, its anti-noise and anti-confusion performances are also better than the conventional deep learning-based methods.

Multi-Aspect-Aware Bidirectional LSTM Networks for Synthetic Aperture Radar Target Recognition

As for the problem of too long training time of convolution neural network (CNN), this paper proposes a fast training method for CNN in SAR automatic target recognition (ATR). The CNN is divided into two parts: one that contains all the convolution layers and sub-sampling layers is considered as convolutional auto-encoder (CAE) for unsupervised training to extract high-level features; the other that contains fully connected layers is regarded as shallow neural network (SNN) to work as a classifier. The experiment based on MSATR database shows that the proposed method can tremendously reduce the training time with little loss of recognition rate.

SAR ATR based on dividing CNN into CAE and SNN

The invention discloses a multi-view SAR image target recognition method based on a depth neural network, and the method comprises three steps: image preprocessing, feature extraction based on CAE, and multi-view SAR image recognition based on RNN. The method specifically comprises the steps: firstly carrying out the cutting and energy normalization of an inputted image; secondly extracting the features of an original image through the non-supervision training of CAE; thirdly constructing a multi-view SAR image feature sequence through the above features; fourthly carrying out the supervised training of the RNN through employing a training feature sequence, wherein the RNN can be used for the recognition of a testing set feature sequence after training. The method can make the most of the capability of CNN in learning and extracting the general features of the image and the capability of RNN in fully extracting the context of the sequence, effectively improves the recognition rate of a multi-view SAR image target, and is higher in engineering value.

Multi-view SAR image target recognition method based on depth neural network

As for the lack of the contribution by decision fusion in pose estimation and the demand for the combination of the feature fusion and the decision fusion in SAR ATR, in this paper, with the help of pose estimation, a new multi-look SAR ATR method is proposed in order to improve the performance, which is based on two-level decision fusion of neural network and sparse representation. The first-level decision fusion is acted for the combination of the pose estimation result by neural network and sparse representation. Based on the constraint of pose, these two models are exerted for the multi-look SAR ATR, and the second-level decision fusion is used to achieve the final recognition result. Several experiments based on MSTAR are conducted, and experimental results show that our method can achieve an acceptable result.

Xuan Li

Papers

SAR ATR based on dividing CNN into CAE and SNN

Multi-view SAR image target recognition method based on depth neural network

Multi-Look SAR ATR Using Two-Level Decision Fusion of Neural Network and Sparse Representation