Small data hold big artificial intelligence (AI) potential. As one of the promising small data AI approaches, few-shot learning has the goal to learn a model efficiently that can recognize novel classes with extremely limited training samples. Therefore, it is critical to accumulate useful prior knowledge obtained from large-scale base class dataset. To realize few-shot scene classification of optical remote sensing images, we start from a baseline model that trains all base classes using a standard cross-entropy loss leveraging two auxiliary objectives to capture intrinsical characteristics across the semantic classes. Specifically, rotation prediction learns to recognize the 2-D rotation of an input to guide the learning of class-transferable knowledge, and contrastive learning aims to pull together the positive pairs while pushing apart the negative pairs to promote intraclass consistency and interclass inconsistency. We jointly optimize two such pretext tasks and semantic class prediction task in an end-to-end manner. To further overcome the overfitting issue, we introduce a regularization technique, adversarial model perturbation, to calibrate the pretext tasks so as to enhance the generalization ability. Extensive experiments on public remote sensing benchmarks including Northwestern Polytechnical University (NWPU)-RESISC45, aerial image dataset (AID), and Wuhan University (WHU)-remote sensing (RS)-19 demonstrate that our method works effectively and achieves best performance that significantly outperforms many state-of-the-art approaches.

Vehicle detection from earth-observation (EO) image has been attracting remarkable attention for its critical value in a variety of applications. Encouraged by the stunning success of deep learning techniques based on convolutional neural networks (CNNs), which have revolutionized the visual data processing community and obtained the state-of-the-art performance in a variety of classification and recognition tasks on benchmark datasets, we propose a network, called EOVNet (EO image-based vehicle detection network), to bridge the gap between the advanced deep learning research of object detection and the specific task of vehicle detection in EO images. Our network has integrated nearly all advanced techniques including very deep residual networks for feature extraction, feature pyramid to fuse multiscale features, network for proposal generation with feature sharing, and hard example mining. Moreover, our novel designs of probability-based localization and homography-based data augmentation have been investigated, resulting in further improvement of the detection performance. For performance evaluation, we have collected nearly all the representative EO datasets associated with vehicle detection. Extensive experiments on the representative datasets demonstrate that our method outperforms the state-of-the-art object detection approach Faster R-CNN++ (which is based on the Faster R-CNN framework, but with significant improvement) with 5% average precision improvement. The source code will be made available after the review process.

Although advanced single image deraining methods have been proposed, their generalization ability to real-world images is usually limited, especially when dealing with rain patterns of different densities, shapes, and directions. In order to improve the robustness and generalization of these deraining methods, we propose a novel density-aware single image deraining method with gated multi-scale feature fusion, which consists of two stages. In the first stage, a sophisticated physics model is leveraged for initial deraining and a network branch is utilized for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), attempting to eliminate artifacts and improve the restoration quality. Extensive experiments have been conducted on the representative synthetic rain datasets and real rain scenes, demonstrating the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-arts.

The proposed method inserts a watermark into the spectral components of the data using techniques analogous to spread spectrum communications, hiding a narrow band signal in a wideband channel. The watermark is difficult for an attacker to remove. It is also robust to common signal and geometric distortion such as JPEG compression, cropping and scaling. In addition, the watermark can be extracted accurately without source host image at the receiver.