Background / introduction: SAR image automatic target recognition technology (SAR-ATR) is one of the research hotspots in the field of image cognitive learning. Inspired by the human cognitive process, experts have designed convolutional neural networks (CNN) based methods and successfully applied the methods to SAR-ATR. However, the performance of CNNs significantly deteriorates when the labelled samples are insufficient. Methods: To effectively utilize the unlabelled samples, a semi-supervised CNN method is proposed in this paper. First, CNN is used to extract the features of the samples, and subsequently the class probabilities of the unlabelled samples are computed using the softmax function. To improve the effectiveness of the unlabelled samples, we remove possible noise performing thresholding on the class probabilities. Afterwards, based on the remaining class probabilities, the information contained in the unlabelled samples is integrated with the scatter matrices of the standard linear discriminant analysis (LDA) method. The loss function of CNN consists of a supervised component and an unsupervised component, where the supervised component is created using the cross-entropy function and the unsupervised component is created using the scatter matrices. The class probabilities are utilized to control the impact of the unlabelled samples in the training process, and the reliability of the unlabelled samples is further improved. Results: We choose ten types of targets from the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset. The experimental results show that the recognition accuracy of 2 our method is significantly higher than that of the supervised CNN method. Conclusions: It proves that our method can effectively improve the SAR-ATR accuracy despite the deficiency of the labelled samples.

Convolutional neural network (CNN) can be applied in synthetic aperture radar (SAR) object recognition for achieving good performance. However, it requires a large number of the labelled samples in its training phase, and therefore its performance could decrease dramatically when the labelled samples are insufficient. To solve this problem, in this paper, we present a novel active semisupervised CNN algorithm. First, the active learning is used to query the most informative and reliable samples in the unlabelled samples to extend the initial training dataset. Next, a semisupervised method is developed by adding a new regularization term into the loss function of CNN. As a result, the class probability information contained in the unlabelled samples can be maximally utilized. The experimental results on the MSTAR database demonstrate the effectiveness of the proposed algorithm despite the lack of the initial labelled samples.

As an important step of synthetic aperture radar image interpretation, synthetic aperture radar image segmentation aims at segmenting an image into different regions in terms of homogeneity. Because of the deficiency of the labeled samples and the existence of speckling noise, synthetic aperture radar image segmentation is a challenging task. We present a new method for synthetic aperture radar image segmentation in this paper. Due to the large size of the original synthetic aperture radar image, we first divide the input image into small slices. Then the image slices are input to the attention based fully convolutional network for obtaining the segmentation results. Finally, the fully connected conditional random field is adopted for improving the segmentation performance of the network. The innovations of our method are as follows: (1) The attention based fully convolutional network is embedded with the multi-scale attention network which is capable of enhancing the extraction of the image features through three strategies: multi-scale feature extraction, channel attention extraction, and spatial attention extraction. (2) We design a new loss function for the attention fully convolutional network by combining lovasz-softmax and cross-entropy losses. The new loss allows us to simultaneously optimize the Intersection over Union and the pixel classification accuracy of the segmentation results. The experiments are performed on two airborne synthetic aperture radar image databases. It has been proved that our method is superior to other state of the art image segmentation approaches.

No abstract

The traditional classification method based on supervised learning classifies remote sensing (RS) images by using sufficient labeled samples. However, the number of labeled samples is limited due to the expensive and time-consuming collection. To effectively utilize the information of unlabeled samples in the learning process, this paper proposes a novel semi-supervised classification method based on class certainty of samples (CCS). First, the class certainty of unlabeled samples obtained based on multi-class SVM is smoothed for robustness. Then, a new semisupervised linear discriminant analysis (LDA) is presented based on class certainty, which improves the separability of samples in the projection subspace. Ultimately, we extend the semi-supervised LDA to nonlinear dimensional reduction by combining class certainty and kernel methods. Furthermore, to assess the effectiveness of proposed method, the nearest neighbor classifier is adopted to classify actual SAR images. The results demonstrate that the proposed method can effectively exploit the information of unlabeled samples and greatly improve the classification effect compared with other state-of-the-art approaches.