Convolutional neural networks (CNNs) have demonstrated their ability object detection of very high resolution remote sensing images. However, CNNs have obvious limitations for modeling geometric variations in remote sensing targets. In this paper, we introduced a CNN structure, namely deformable ConvNet, to address geometric modeling in object recognition. By adding offsets to the convolution layers, feature mapping of CNN can be applied to unfixed locations, enhancing CNNs’ visual appearance understanding. In our work, a deformable region-based fully convolutional networks (R-FCN) was constructed by substituting the regular convolution layer with a deformable convolution layer. To efficiently use this deformable convolutional neural network (ConvNet), a training mechanism is developed in our work. We first set the pre-trained R-FCN natural image model as the default network parameters in deformable R-FCN. Then, this deformable ConvNet was fine-tuned on very high resolution (VHR) remote sensing images. To remedy the increase in lines like false region proposals, we developed aspect ratio constrained non maximum suppression (arcNMS). The precision of deformable ConvNet for detecting objects was then improved. An end-to-end approach was then developed by combining deformable R-FCN, a smart fine-tuning strategy and aspect ratio constrained NMS. The developed method was better than a state-of-the-art benchmark in object detection without data augmentation.

Change detection (CD) is an essential earth observation technique. It captures the dynamic information of land objects. With the rise of deep learning, convolutional neural networks (CNN) have shown great potential in CD. However, current CNN models introduce backbone architectures that lose detailed information during learning. Moreover, current CNN models are heavy in parameters, which prevents their deployment on edge devices such as UAVs. In this work, we tackle this issue by proposing RDP-Net: a region detail preserving network for CD. We propose an efficient training strategy that constructs the training tasks during the warmup period of CNN training and lets the CNN learn from easy to hard. The training strategy enables CNN to learn more powerful features with fewer FLOPs and achieve better performance. Next, we propose an effective edge loss that increases the penalty for errors on details and improves the network’s attention to details such as boundary regions and small areas. Furthermore, we provide a CNN model with a brand new backbone that achieves the state-of-the-art empirical performance in CD with only 1.70M parameters. We hope our RDP-Net would benefit the practical CD applications on compact devices and could inspire more people to bring change detection to a new level with the efficient training strategy. The code and models are publicly available at https://github.com/Chnja/RDPNet.

In recent years, deep neural network (DNN) based scene classification methods have achieved promising performance. However, the data-driven training strategy requires a large number of labeled samples, making the DNN-based methods unable to solve the scene classification problem in the case of a small number of labeled images. As the number and variety of scene images continue to grow, the cost and difficulty of manual annotation also increase. Therefore, it is significant to deal with the scene classification problem with only a few labeled samples. In this paper, we propose an attention metric network (AMN) in the framework of the few-shot learning (FSL) to improve the performance of one-shot scene classification. AMN is composed of a self-attention embedding network (SAEN) and a cross-attention metric network (CAMN). In SAEN, we adopt the spatial attention and the channel attention of feature maps to obtain abundant features of scene images. In CAMN, we propose a novel cross-attention mechanism which can highlight the features that are more concerned about different categories, and improve the similarity measurement performance. A loss function combining mean square error (MSE) loss with multi-class N-pair loss is developed, which helps to promote the intra-class similarity and inter-class variance of embedding features, and also improve the similarity measurement results. Experiments on the NWPU-RESISC45 dataset and the RSD-WHU46 dataset demonstrate that our method achieves the state-of-the-art results on one-shot remote sensing image scene classification tasks.

Limited labels and increasing multisource data promote domain adaptation (DA) problem as a challenging study for polarimetric synthetic aperture radar (PolSAR) interpretation. Existing DAs for optical images cannot generalize over PolSAR imagery due to its special side-imaging characteristics and complex distribution shifts. In this letter, a general feature paradigm (GFP) is proposed for unsupervised cross-domain PolSAR image classification. The GFP is based on a key observation that interclass aggregation is optimized after four-step feature transformations. This key observation leads to GFP that not only reduces the domain shifts but also compatible with typical DA methods. The GFPs are conducted on both source and target domain by unsupervised manner, including polarimetric basis extraction, the Wishart clustering, histogram statistics, and dimensionality reduction. After these transformations, the unlabeled target PolSAR image can be classified based on obtained GFP, DA, and limited labeled samples only from the source domain. Extensive unsupervised cross-domain experiments on 27 scenarios verified that GFP leads to at most 93.76% accuracy for full- and dual-polarized synthetic aperture radar (SAR) images’ classification. Moreover, the GFP shed light on extensive cross-domain PolSAR applications about built-up areas, vegetation, and bare land analysis.

Increasing amounts of polarimetric synthetic aperture radar (PolSAR) images from different sensors covering different scenes are available, but limited labeled samples and trained models can hardly work well in these cross-domain data interpretations. Fortunately, domain adaptation (DA) can transfer knowledge in existing images to new yet related images. DA shows attractive potential for PolSAR classification, and it is still challenging due to more complex domain shifts caused by different sensors, imaging conditions, and distributions. Inspired by the widely applicable polarimetric scattering mechanisms and DA ability of subspace alignment (SA), this article is devoted to constructing a robust unsupervised cross-domain PolSAR classification framework, by exploring scattering and statistical characteristics mapping between the source and target domains. First, classical scattering components of both source and target data were extracted, and Wishart clustering was adopted to derive the statistical information of scattering components at patch level. Second, the intrinsic polarimetric scattering components were estimated and extracted, which were called statistical scattering components (SSCs). Third, by applying SA, the source SSC was aligned with target SSC, and domain shift was further reduced. Finally, the target PolSAR image was classified based on labeled samples from source domain, and unsupervised cross-domain classification was achieved by SSC-based SA (SSC-SA). The unsupervised cross-domain experiments are conducted on 49 units among 11 data sets, including Radarsat-2, Gaofen-3, AIRSAR, and Pi-SAR images. With randomly selected labeled samples (about 2%–10%) from source domain, the accuracies of the proposed cross-domain classifications range between 80.20% and 95.64%. Also, the proposed SSC feature pattern is proved extensible for other polarimetric basis and decompositions.