Image classification is a fundamental task in remote sensing image processing. In recent years, deep convolutional neural networks (DCNNs) have experienced significant breakthroughs in natural image recognition. The remote sensing field, however, is still lacking a large-scale benchmark similar to ImageNet. In this paper, we propose a remote sensing image classification benchmark (RSI-CB) based on massive, scalable, and diverse crowdsourced data. Using crowdsourced data, such as Open Street Map (OSM) data, ground objects in remote sensing images can be annotated effectively using points of interest, vector data from OSM, or other crowdsourced data. These annotated images can, then, be used in remote sensing image classification tasks. Based on this method, we construct a worldwide large-scale benchmark for remote sensing image classification. This benchmark has large-scale geographical distribution and large total image number. It contains six categories with 35 sub-classes of more than 24,000 images of size $256\times 256$ pixels. This classification system of ground objects is defined according to the national standard of land-use classification in China and is inspired by the hierarchy mechanism of ImageNet. Finally, we conduct numerous experiments to compare RSI-CB with the SAT-4, SAT-6, and UC-Merced data sets. The experiments show that RSI-CB is more suitable as a benchmark for remote sensing image classification tasks than other benchmarks in the big data era and has many potential applications.

The spatial resolution of remote sensing images is continuously improved by the development of remote sensing satellite and sensor technology. Hence, background information in an image becomes increasingly complex and causes considerable interference to the object detection task. Can we pay as much attention to the object in an image as human vision does? This letter proposes a multi-scale spatial and channel-wise attention (MSCA) mechanism to answer this question. MSCA has two advantages that help improve object detection performance. First, attention is paid to the spatial area related to the foreground, and compared with other channels, more attention is given to the feature channel with a greater response to the foreground region. Second, for objects with different scales, MSCA can generate an attention distribution map that integrates multi-scale information and applies it to the feature map of the deep network. MSCA is a flexible module that can be easily embedded into any object detection model based on deep learning. With the attention exerted by MSCA, the deep neural network can efficiently focus on objects of different backgrounds and sizes in remote sensing images. Experiments show that the mean average precision of object detection is improved after the addition of MSCA to the current object detection model.

The lack of pixel-level labeling limits the practicality of deep learning-based building semantic segmentation. Weakly supervised semantic segmentation based on image-level labeling results in incomplete object regions and missing boundary information. This paper proposes a weakly supervised semantic segmentation method for building detection. The proposed method takes the image-level label as supervision information in a classification network that combines superpixel pooling and multi-scale feature fusion structures. The main advantage of the proposed strategy is its ability to improve the intactness and boundary accuracy of a detected building. Our method achieves impressive results on two 2D semantic labeling datasets, which outperform some competing weakly supervised methods and are close to the result of the fully supervised method.

Buildings represent the most relevant features of human activity in urban regions, but their change detection using very-high-resolution (VHR) remote sensing imagery is still a major challenge. Effective representation of the building is the key point in building change detection. The linear feature can indirectly represent the structure and distribution of man-made objects. Thus, this study proposes a shape feature-based building change detection method. Specifically, a line-constrained shape (LCS) feature is developed to capture the shape characteristics of buildings. This feature improves the discriminability between buildings and other ground objects by integrating the pixel shape feature and line segments. The building candidate area (BCA) is created in accordance with the distribution of the line segments in two-phase images. The problem space is constrained in a high-likelihood region of buildings because of the BCA. Comparative experimental results demonstrate that the combination of the spectral feature and the developed LCS feature achieves the best performance in object-based building change detection in VHR imagery.

Multilevel co-location patterns embedded in spatial datasets are difficult to discern due to the complexity of neighboring relationships among spatial features. The neighboring relationships are used to determine whether instances of different spatial features are located in close geographic proximity. When spatial features are distributed unevenly, the neighboring relationships among spatial features cannot be constructed appropriately. Correspondingly, the instances of co-location patterns cannot be generated correctly, and the prevalence of multilevel co-location patterns cannot be measured accurately. To overcome this challenge, this study develops a method to adaptively detect multilevel co-location patterns based on natural neighborhoods. First, locally adaptive neighboring relationships for instances of different spatial features, called ‘natural neighborhoods’, are defined by considering the formation mechanism of co-location patterns and the local-distribution characteristics of spatial features. Using the natural neighborhoods, we propose a multilevel refining method to identify all global and local co-location patterns algorithmically. We compare the proposed method against three state-of-the-art methods using both simulated and real-life datasets. The comparison shows that the proposed method can discover multilevel co-location patterns from unevenly distributed spatial features more completely and accurately with less a priori knowledge for the construction of the natural neighborhoods.

With the development of navigation globalization and ship dehumanization, the contradiction between the increasing demand for ship behavior supervision and limited traffic service resources is obvious, and the frequent occurrence of accidents at sea is a problem. The monitoring of abnormal ship behavior is an important link in maritime transportation. With the popularization of the automatic identification system and increasing research in the maritime field, the AIS is widely used in the management of ship static information and the real-time sharing of dynamic information. The generated moving ship trajectory data provide a new opportunity for research into abnormal ship behavior and its detection. In light of the current situation of abnormal ship behavior research, we detected abnormal ship behavior from the point of view of spatial information and thematic information based on moving ship trajectory data. Therefore, this study first modeled the cognition of abnormal ship behavior. Then, based on the cognition of group ship behavior rules, we used a method based on graph structure learning to mine maritime routes from the perspective of ship spatial information. Next, we used Rayda’s criterion to detect the anomalous behavior of ships in space. Then, based on the isolation forest algorithm, we detected and described the abnormal behavior shown by ship thematic information. The experimental results show that the framework proposed in this paper can effectively detect the abnormal behavior of ships.