Federated Learning (FL) enables multiple institutes to train models collaboratively without sharing private data. Most of the current FL research focuses on perspectives such as communication efficiency, privacy protection, and personalization. Almost all work assumed that the data of FL are already ideally collected. However, in medical image analysis scenarios, data annotation demands both expertise and tedious labor, which means it is a critical problem that cannot be neglected in FL. In this study, we proposed a federated active learning (FedAL) framework that can decrease the annotation workload while maintaining the performance of FL. To the best of our knowledge, this is the first federated active learning framework working on medical images. Using only up to 50% of samples, our FedAL was able to achieve state-of-the-art performance on the real-world dermoscopic task. Our FedAL outperformed active learning methods under FL and achieved the performance comparable to full data FL.

Automatic inshore ship recognition, including target localization and type classification, is an important and challenging problem. However, arbitrarily rotated ships are always moored inshore densely. This makes it very difficult to locate ship targets. To resolve this problem, we proposes a multiclass oriented ship localization and recognition framework based on a cascade region convolutional neural network (R-CNN). First, Cascade R-CNN is adopted to localize and classify the positive regions - a set of bounding boxes (BBox). Second, a novel procedure which transforms a bounding box to a rotated bounding box (B2RB) is designed and applied to each BBox to regress a rotated BBox (RBox) and non-maximum suppression (NMS) is adopted to remove redundant RBoxes. Extensive experimental results conducted on the dataset collected from Google Earth demonstrate the effectiveness of our proposed approach, compared to two other state-of-the-art approaches.

Recently, deep learning-based methods have brought new ideas for ship detection in synthetic aperture radar (SAR) images. However, several challenges still exist: 1) deep models contain millions of parameters, whereas the available annotated samples are not sufficient in number for training. Therefore, most deep detectors have to fine-tune networks pre-trained on ImageNet, which incurs learning bias due to the huge domain mismatch between SAR images and ImageNet images. Furthermore, it has a little flexibility to redesign the network structure; and 2) ships in SAR images are relatively small in size and densely clustered, whereas most deep detectors have poor performance with small objects due to the rough feature map used for detection and the extreme foreground–background imbalance. To address these problems, this paper proposes an effective approach to learn deep ship detector from scratch. First, we design a condensed backbone network, which consists of several dense blocks. Hence, earlier layers can receive additional supervision from the objective function through the dense connections, which makes it easy to train. In addition, feature reuse strategy is adopted to make it highly parameter efficient. Therefore, the backbone network could be freely designed and effectively trained from scratch without using a large amount of annotated samples. Second, we improve the cross-entropy loss to address the foreground–background imbalance and predict multi-scale ship proposals from several intermediate layers to improve the recall rate. Then, position-sensitive score maps are adopted to encode position information into each ship proposal for discrimination. The comparison results on the Sentinel-1 data set show that: 1) learning ship detector from scratch achieved better performance than ImageNet pre-trained model-based detectors and 2) our method is more effective than existing algorithms for detecting the small and densely clustered ships.

Fast multiclass object detection for remote sensing images plays an important role for a wide range of applications. Traditional methods based on a sliding window search lead to heavy computational costs and are unsuitable for multiclass detection. Recently, deep learning algorithms, especially faster region based convolutional neural networks (Faster R-CNN), which adopt a region proposal paradigm to avoid exhaustive search, has achieved state-of-the-art multiclass detection performance in computer vision. This paper investigates the use of Faster R-CNN in the earth observation community. We have three contributions: 1) It's the first time to successfully use Faster R-CNN for object detection in remote sensing images. It achieved faster speed (22 ×faster) and better performance (a mAP of 78% vs. 72%) than traditional methods; 2) we adopt data augmentation to train Faster R-CNN with limited samples; 3) we successfully tested our method on large-scale google earth images, which shows robustness of our method.

This paper proposes a coupled region based convolutional neural networks (R-CNN) to automatically detect vehicles in aerial images. Traditional methods are mostly based on sliding-window search, and use handcrafted or shallow-learning based features. They have limited description ability and heavy computational costs. Recently, a series of R-CNN based methods have achieved great success in general object detection. Inspired by the previous work, we propose a coupled R-CNN to detect small size vehicles in large-scale aerial images. First, a vehicle proposal network (VPN) is proposed to generate candidate vehicle-like regions, using a hyper feature map combined by feature maps of different layers. Then, a vehicle classification network (VCN) is developed to further verify the candidate regions and classify vehicles in eight directions. In this study, our method is tested on a challenge Munich vehicle dataset and the collected vehicle dataset, with improvements in accuracy and speed compared to existing methods.

Partial configuration model (PCM) is an occluded object detection method in high-resolution remote sensing images (HR-RSIs) based on the deformable part-based model (DPM). However, it needs extra category predefinition, considerable partlevel annotation, and repeated multimodel training. In this letter, an automatic and fast PCM generation method is proposed based on a novel part sharing mechanism. We propose to share parts from one trained DPM model (tDPM) among different models of partial configurations (PCs) to address the above problems. PCs are first designed according to part anchors of tDPM. The model is then generated through corresponding parts selection, root coverage cropping, and elements reweighing. This method avoids the need for manual category predefinition and partlevel annotation, while largely reducing the computation of PCM training. Experimental results on three HR-RSI data sets show that the proposed method obtains a training speedup of 6.7× and 2× for each PC of airplane and ship categories, while achieving a comparable accuracy compared with PCM.

In this paper, we address the problem of semi-supervised visual domain adaptation for transferring scene category models from ground view images to overhead view very high-resolution (VHR) remote sensing images. We introduce a multiple kernel learning domain adaptation algorithm to fuse the information from multiple features and cope with the considerable variation in feature distributions between images from two domains. For each image, we first extract eight state-of-art local features and use the pretrained scene attribute model from ground-level SUN attribute database to predict attribute labels. For each scene class we learn an adapted target classifier based on multiple feature kernels by minimizing both the structural risk functional and the mismatch between data distributions of two domains. Experimental results demonstrate that it is possible to use a scene category model learned on a set of ground view scenes for semi-supervised classification of VHR remote sensing images.

This paper aims to address the problem of unsupervised scene model adaptation for transferring semantic labels from ground level images to aerial and satellite scenes. Specifically, we present a novel unsupervised domain adaptation algorithm based on subspace alignment. The core idea is to reduce the feature distribution discrepancy between ground view images and overhead view remote sensing images in a latent discriminative subspace. We first generate pseudo-labels for the remote sensing data by applying spectral clustering to a cross-domain similarity matrix, which is built from sparse coefficients found in a low-dimensional latent space. This coarse alignment between the two views exploits the assumption that the collection of data of different classes from both domains can be viewed as samples from a union of low-dimensional subspaces. Then, we create discriminative subspaces for both domains using partial least squares correlation. Finally, a mapping which aligns the discriminative source subspace into the target one is learned by minimizing a Bregman matrix divergence function. Experimental results on aerial-to-satellite, ground-to-aerial and ground-to-satellite scene image data sets demonstrate that the proposed method outperforms the baselines and several state-of-the-art competing methods.

Speckle existed in SAR image is an undesirable product of specific imaging principle which influences SAR image interpretation and processing. In this paper, a new SAR image denoising algorithm has been proposed combining cluster with sparse representation under the non-local methodology. Due to the similar clustered patches, the sparsity coding of clustered patches is sparser. And clustered patches with similar structure could have the same constraint condition defined by the center of clustering. Thus, the non-local patches are clustered and filtered as a whole with shrinked sparsity coding. This algorithm has preferable denoising results on both simulated images and real SAR images. Experiments show prospects with speckle of different degrees compared with state-of-the-art despeckling methods. Proposed algorithm performs well both in noise reduction and detail preservation.

In Cognitive Radio (CR) networks, it is common that the spectrum holes are too narrow to support high-speed communications. Discontinuous Orthogonal Frequency Division Multiplexing (DOFDM) is a good way for a secondary user to access several spectrum fragments simultaneously with one Radio Front (RF). In this paper, a novel Channel Characteristic Aware Spectrum Aggregation (CCASA) algorithm which uses DOFDM to aggregation spectrum fragments with only one radio front is proposed in order to increase the overall throughput of a CR network. By combining Adaptive Modulation and Coding (AMC) and spectrum aggregation, the good subcarriers are assigned to the specific secondary users in CCASA algorithm thus achieving a better channel efficiency. Different bandwidth requirement and aggregation limitation of secondary users are both considered in this algorithm while maintaining a fairly low computational complexity. The simulation results show that CCASA achieves a bigger total throughput than existing aggregation algorithms.