20 May 2022 in IEEE Transactions on Neural Networks and Learning Systems
As a typical label-limited task, it is significant and valuable to explore networks that enable to utilize labeled and unlabeled samples simultaneously for synthetic aperture radar (SAR) image scene classification. Graph convolutional network (GCN) is a powerful semisupervised learning paradigm that helps to capture the topological relationships of scenes in SAR images. While the performance is not satisfactory when existing GCNs are directly used for SAR image scene classification with limited labels, because few methods to characterize the nodes and edges for SAR images. To tackle these issues, we propose a contrastive learning-based dual dynamic GCN (DDGCN) for SAR image scene classification. Specifically, we design a novel contrastive loss to capture the structures of views and scenes, and develop a clustering-based contrastive self-supervised learning model for mapping SAR images from pixel space to high-level embedding space, which facilitates the subsequent node representation and message passing in GCNs. Afterward, we propose a multiple features and parameter sharing dual network framework called DDGCN. One network is a dynamic GCN to keep the local consistency and nonlocal dependency of the same scene with the help of a node attention module and a dynamic correlation matrix learning algorithm. The other is a multiscale and multidirectional fully connected network (FCN) to enlarge the discrepancies between different scenes. Finally, the features obtained by the two branches are fused for classification. A series of experiments on synthetic and real SAR images demonstrate that the proposed method achieves consistently better classification performance than the existing methods.
5 May 2021 in IEEE Transactions on Geoscience and Remote Sensing
Scatter targets of complex land covers in polarimetric synthetic aperture radar (PolSAR) images are often randomly oriented and cause randomly fluctuating echoes, which brings a challenge to PolSAR image classification. Therefore, many existing methods have alleviated this problem through orientation compensation. However, there are still two obstacles that limit the improvement of classification accuracy. On the one hand, generally, these methods process PolSAR images with fixed polarization rotation angles, which is experience-dependent and inflexible. On the other hand, for the different land covers of a PolSAR image, the existing methods do not consider these rotation angles separately. For the first obstacle, we design a group of convolution kernels called polarization rotation kernels (PRKs) and utilize them to build the polarimetric convolutional neural network (CNN) (PolCNN). The PolCNN is the base network of our final model, and it can learn polarization rotation angles adaptively. For the second obstacle, we extend the PolCNN into a multipath structure, the final model polarimetric multipath CNN (PolMPCNN). The polarization rotation angles of different land covers are directly related to the networks of different paths within the PolMPCNN. Furthermore, we also put forward the two-scale sampling and the stagewise training algorithm in order that our PolMPCNN can fit different scales of PolSAR targets and pays more attention to difficult training samples. Experiments on real PolSAR images show that the proposed model achieves the best classification results with an extremely low sampling rate of 0.1%.
21 January 2021 in IEEE Transactions on Geoscience and Remote Sensing
With powerful feature representations, convolutional neural networks (CNNs) have produced tremendous achievements in image classification tasks and, typically, entail millions of labeled samples to train massive parameters. However, the sample labeling of synthetic aperture radar (SAR) images is extremely difficult, especially pixelwise labels, and has, sometimes, required field trips to accomplish labeling. Moreover, the inherent speckle noise may weaken the ability of networks to extract effective features from SAR images. In this article, we address these issues by labeling a few patchwise samples and propose Ridgelet-Nets with speckle reduction regularization for SAR image scene classification by combining deep learning with multiscale geometric analysis and statistical modeling of SAR images. First, we design Ridgelet-Nets with convolutional kernels constructed by ridgelet filters to reduce the training parameters and learn more discriminative features. Then, we embed speckle reduction regularization in the Ridgelet-Nets to restrain the influence of speckle noise and smooth the classification maps, in which the prior information of SAR image statistical modeling is introduced. Finally, we propose an adaptive SAR image scene classification framework based on an extended hierarchical visual semantic model, considering the differences in the structures and spatial relationships of different regions in the SAR images, particularly large-scale and complex scenes. Experimental results on real SAR images demonstrate that the proposed framework can achieve preferable classification performance using very limited labeled samples.
26 February 2021 in IEEE Transactions on Geoscience and Remote Sensing
Data-based image classification methods, such as convolutional neural networks (CNNs), have achieved state-of-the-art performance. They usually leverage thousands of labeled samples to train the networks but ignore some prior knowledge. However, labeled samples are difficult to be obtained for synthetic aperture radar (SAR) images. Model-based methods are adept at utilizing the prior information of data, while they have to introduce some restrictions or assumptions during the realization of models. Consequently, to develop the advantages of both methods and improve their disadvantages, we propose a hybrid network by coupling the data-based with model-based methods for SAR image scene classification in this article. First, to fully use the prior information of SAR images and large amounts of unlabeled samples, we improve the $G^{0}$ -based variational Bayesian inference model (GVBI) and construct a $G^{0}$ -based convolutional variational auto-encoder (GCVAE) for unsupervised learning of the distributional characteristics of SAR images. After that, we further extend the GCVAE by combining it with CNN, resulting in a stronger hybrid network to classify SAR images with a few labeled samples. In addition, considering the abundant structural information is crucial for SAR image classification, we design a sketch fitter and two structural constraints on both pixel and sketch spaces to assist the hybrid network to improve its classification performance. Finally, we evaluate the performance of our method on real-SAR images, and the experimental results demonstrate that the proposed framework outperforms related methods on classification while reducing the manual annotation substantially.