Object detection has attracted increasing attention in the field of remote sensing image analysis. Complex backgrounds, vertical views, and variations in target kind and size in remote sensing images make object detection a challenging task. In this work, considering that the types of objects are often closely related to the scene in which they are located, we propose a convolutional neural network (CNN) by combining scene-contextual information for object detection. Specifically, we put forward the scene-contextual feature pyramid network (SCFPN), which aims to strengthen the relationship between the target and the scene and solve problems resulting from variations in target size. Additionally, to improve the capability of feature extraction, the network is constructed by repeating a building aggregated residual block. This block increases the receptive field, which can extract richer information for targets and achieve excellent performance with respect to small object detection. Moreover, to improve the proposed model performance, we use group normalization, which divides the channels into groups and computes the mean and variance for normalization within each group, to solve the limitation of the batch normalization. The proposed method is validated on a public and challenging dataset. The experimental results demonstrate that our proposed method outperforms other state-of-the-art object detection models.

With exploiting contextual information over large image regions in an efficient way, the deep convolutional neural network has shown an impressive performance for single-image super-resolution (SR). In this paper, we propose a new deep convolutional network by cascading multiple well-designed inception-residual blocks within the deep Laplacian pyramid framework to progressively restore the missing high-frequency details in the low-resolution images. By optimizing our network structure, the trainable depth of our proposed network gains a significant improvement, which in turn improves super-resolving accuracy. However, the saturation and degradation of training accuracy remains a critical problem. With regard to this, we propose an effective two-stage training strategy, in which we first use the images downsampled from the ground-truth high-resolution (HR) images to pretrain the inception-residual blocks on each pyramid level with an extremely high learning rate enabled by gradient clipping, and then the original ground-truth HR images are used to fine-tune all the pretrained inception-residual blocks for obtaining our final SR models. Furthermore, we present a new loss function operating in both image space and local rank space to optimize our network for exploiting the contextual information among different output components. Extensive experiments on benchmark data sets validate that the proposed method outperforms the existing state-of-the-art SR methods in terms of the objective evaluation as well as the visual quality.

How to efficiently utilize vast amounts of easily accessed aerial imageries is a critical challenge for researchers with the proliferation of high-resolution remote sensing sensors and platforms. Recently, the rapid development of deep neural networks (DNN) has been a focus in remote sensing, and the networks have achieved remarkable progress in image classification and segmentation tasks. However, the current DNN models inevitably lose the local cues during the downsampling operation. Additionally, even with skip connections, the upsampling methods cannot properly recover the structural information, such as the edge intersections, parallelism, and symmetry. In this paper, we propose the Web-Net, which is a nested network architecture with hierarchical dense connections, to handle these issues. We design the Ultra-Hierarchical Sampling (UHS) block to absorb and fuse the inter-level feature maps to propagate the feature maps among different levels. The position-wise downsampling/upsampling methods in the UHS iteratively change the shape of the inputs while preserving the number of their parameters, so that the low-level local cues and high-level semantic cues are properly preserved. We verify the effectiveness of the proposed Web-Net in the Inria Aerial Dataset and WHU Dataset. The results of the proposed Web-Net achieve an overall accuracy of 96.97% and an IoU (Intersection over Union) of 80.10% on the Inria Aerial Dataset, which surpasses the state-of-the-art SegNet 1.8% and 9.96%, respectively; the results on the WHU Dataset also support the effectiveness of the proposed Web-Net. Additionally, benefitting from the nested network architecture and the UHS block, the extracted buildings on the prediction maps are obviously sharper and more accurately identified, and even the building areas that are covered by shadows can also be correctly extracted. The verified results indicate that the proposed Web-Net is both effective and efficient for building extraction from high-resolution remote sensing images.

The automated building detection in aerial images is a fundamental problem encountered in aerial and satellite images analysis. Recently, thanks to the advances in feature descriptions, Region-based CNN model (R-CNN) for object detection is receiving an increasing attention. Despite the excellent performance in object detection, it is problematic to directly leverage the features of R-CNN model for building detection in single aerial image. As we know, the single aerial image is in vertical view and the buildings possess significant directional feature. However, in R-CNN model, direction of the building is ignored and the detection results are represented by horizontal rectangles. For this reason, the detection results with horizontal rectangle cannot describe the building precisely. To address this problem, in this paper, we proposed a novel model with a key feature related to orientation, namely, Oriented R-CNN (OR-CNN). Our contributions are mainly in the following two aspects: 1) Introducing a new oriented layer network for detecting the rotation angle of building on the basis of the successful VGG-net R-CNN model; 2) the oriented rectangle is proposed to leverage the powerful R-CNN for remote-sensing building detection. In experiments, we establish a complete and bran-new data set for training our oriented R-CNN model and comprehensively evaluate the proposed method on a publicly available building detection data set. We demonstrate State-of-the-art results compared with the previous baseline methods.

With the development of deep learning technology, an enormous number of convolutional neural network (CNN) models have been proposed to address the challenging building extraction task from very high-resolution (VHR) remote sensing images. However, searching for better CNN architectures is time-consuming, and the robustness of a new CNN model cannot be guaranteed. In this paper, an improved boundary-aware perceptual (BP) loss is proposed to enhance the building extraction ability of CNN models. The proposed BP loss consists of a loss network and transfer loss functions. The usage of the boundary-aware perceptual loss has two stages. In the training stage, the loss network learns the structural information from circularly transferring between the building mask and the corresponding building boundary. In the refining stage, the learned structural information is embedded into the building extraction models via the transfer loss functions without additional parameters or postprocessing. We verify the effectiveness and efficiency of the proposed BP loss both on the challenging WHU aerial dataset and the INRIA dataset. Substantial performance improvements are observed within two representative CNN architectures: PSPNet and UNet, which are widely used on pixel-wise labelling tasks. With BP loss, UNet with ResNet101 achieves 90.78% and 76.62% on IoU (intersection over union) scores on the WHU aerial dataset and the INRIA dataset, respectively, which are 1.47% and 1.04% higher than those simply trained with the cross-entropy loss function. Additionally, similar improvements (0.64% on the WHU aerial dataset and 1.69% on the INRIA dataset) are also observed on PSPNet, which strongly supports the robustness of the proposed BP loss.

Object detection is one of the core technologies in aerial image processing and analysis. Although existing aerial image object detection methods based on deep learning have made some progress, there are still some problems remained: (1) Most existing methods fail to simultaneously consider multi-scale and multi-shape object characteristics in aerial images, which may lead to some missing or false detections; (2) high precision detection generally requires a large and complex network structure, which usually makes it difficult to achieve the high detection efficiency and deploy the network on resource-constrained devices for practical applications. To solve these problems, we propose a slimmer network for more efficient object detection in aerial images. Firstly, we design a polymorphic module (PM) for simultaneously learning the multi-scale and multi-shape object features, so as to better detect the hugely different objects in aerial images. Then, we design a group attention module (GAM) for better utilizing the diversiform concatenation features in the network. By designing multiple detection headers with adaptive anchors and the above-mentioned two modules, we propose a one-stage network called PG-YOLO for realizing the higher detection accuracy. Based on the proposed network, we further propose a more efficient channel pruning method, which can slim the network parameters from 63.7 million (M) to 3.3M that decreases the parameter size by 94.8%, so it can significantly improve the detection efficiency for real-time detection. Finally, we execute the comparative experiments on three public aerial datasets, and the experimental results show that the proposed method outperforms the state-of-the-art methods.