Convolutional neural network (CNN)-based methods are widely used in remote sensing image scene classification and can obtain excellent performances. However, the stacked receptive fields in the CNN-based methods have limitations in modeling the long-range dependencies of local features. The vision transformer (ViT) model provides a good solution as it directly considers the global interactions of local patches by the self-attention mechanism. However, the vanilla ViT model, which simply splits images into fixed-size patches treated as tokens, mainly considers the global information in the spatial domain. In this article, a spatial-channel feature preserving ViT (SCViT) model is proposed, which considers both the detailed geometric information of the high-spatial-resolution (HSR) imagery and the contribution of the different channels contained in the classification token. First, in the proposed method, tokens are generated by progressively aggregating the neighboring overlapping patches to extract the local structural features of the imagery. Second, a multihead self-attention (MSA) mechanism is used to model the global interactions of the tokens in the encoder. A lightweight channel attention (LCA) module is then introduced to consider the importance of the different channels in the classification token. Finally, a multilayer perceptron (MLP) is used to acquire the final results. Compared with the state-of-the-art scene classification methods, the experimental results confirm the potential of using ViT models in remote sensing image scene classification.

Landslides pose a serious threat to human life, safety, and natural resources. Remote sensing images can be used to effectively monitor landslides at a large scale, which is of great significance for pre-disaster warning and post-disaster assistance. In recent years, deep learning based methods have made great progress in the field of remote sensing image landslide detection. In remote sensing images, landslides display a variety of scales and shapes. In this paper, to better extract and keep the multi-scale shape information of landslides, a shape-enhanced vision transformer (ShapeFormer) model is proposed. For the feature extraction, a pyramid vision transformer (PVT) model is introduced, which directly models the global information of local elements at different scales. To learn the shape information of different landslides, a shape feature extraction branch is designed, which uses the adjacent feature maps at different scales in the PVT model to improve the boundary information. After the feature extraction step, a decoder with deconvolutional layers follows, which combines the multiple features and gradually recovers the original resolution of the combined features. A softmax layer is connected with the combined features to acquire the final pixel-wise result. The proposed ShapeFormer model was tested on two public datasets—the Bijie dataset and the Nepal dataset—which have different spectral and spatial characteristics. The results, when compared with those of some of the state-of-the-art methods, show the potential of the proposed method for use with multi-source optical remote sensing data for landslide detection.

High spatial resolution (HSR) remote sensing images provide detailed geometric information about land cover. As a result, it is possible to detect more subtle changes with the help of HSR images. However, due to the increased spatial resolution and the limited spectral information, it is difficult to identify the real changes only through the spectral feature of the image. To fully explore the spectral-spatial information and improve the change detection performance for HSR images, this paper proposes the hybrid conditional random field (HCRF) model, which combines the traditional random field method with an object-based technique. In the proposed method, the spectral discriminative information of a single pixel is extracted by the unary potential, which is modeled using a soft clustering method to make an initial separation of changed and unchanged pixels. The pairwise potential then considers the contextual information of adjacent pixels to favor spatial smoothing. An object term is also introduced in the HCRF model to keep the homogeneity of changed objects. By the use of these approaches, the oversmoothing problem of the random field-based methods and the detection error caused by the segmentation strategy in the object-based methods can be relieved. The proposed method was tested on three HSR image data sets and outperformed the compared state-of-the-art techniques.

Satellite video is an emerging data source for dynamic Earth observation, which provides us with a new means for large-scale moving vehicle detection and traffic monitoring. However, the ratio of the foreground to background in satellite video is severely uneven. Furthermore, due to the high altitude and spatial resolution of video satellites, the moving vehicles often have a small size and low contrast with the background. There is also complex environment and background movement due to the imaging mode of satellite video, which makes the vehicle detection more challenging. In this paper, to solve the problems of the low contrast and the dynamic background in satellite video, a local enhancement fusion Gaussian mixture model and improved three-frame difference detection framework, namely E-GMMTFD, is proposed to detect small moving vehicles in satellite video. In the E-GMMTFD framework, the local contrast enhancement method is used to improve the contrast between the target and the background. The Gaussian mixture model (GMM) and the three-frame difference (TFD) method are fused by a logical operation to detect moving vehicles. The E-GMMTFD framework utilizes the adaptability to complex environment changes of the GMM and the strength of the TFD method to eliminate background movement, thus reducing the false alarms caused by illumination change and background movement. Three experiments were performed on moving vehicle detection using Jilin-1 satellite video and SkySat-1 satellite video. The experimental results confirmed that the proposed framework can be efficiently used to detect moving vehicles in the large scale of satellite video. Based on the vehicle detection results for the satellite video, the traffic density of the urban road network can provide technical support for alleviating urban traffic congestion.

In the recent years, convolutional neural networks (CNN)-based super resolution (SR) methods are widely used in the field of remote sensing. However, complicated remote sensing images contain abundant high-frequency details, which are difficult to capture and reconstruct effectively. To address this problem, we propose a dense channel attention network (DCAN) to reconstruct high-resolution (HR) remote sensing images. The proposed method learns multi-level feature information and pays more attention to the important and useful regions in order to better reconstruct the final image. Specifically, we construct a dense channel attention mechanism (DCAM), which densely uses the feature maps from the channel attention block via skip connection. This mechanism makes better use of multi-level feature maps which contain abundant high-frequency information. Further, we add a spatial attention block, which makes the network have more flexible discriminative ability. Experimental results demonstrate that the proposed DCAN method outperforms several state-of-the-art methods in both quantitative evaluation and visual quality.

Recently, transformer models have been introduced into the field of remote sensing image object detection, benefiting from their ability to model long-term information. However, the existing transformer-based object detection methods mainly consider the global interaction of local elements and have a limited ability to enhance the local information, which can bring some difficulties in distinguishing real objects and a complex background. In this letter, a query-enhanced transformer (QETR) model is proposed to solve the above problems. The proposed model consists of three main parts: an encoder, a decoder, and a detection head. A Swin transformer is used to extract deep features in the encoder. In the decoder, the object and anchor queries are initialized and the feature and position information of the objects is learned by the multi-head self-attention and cross-attention mechanisms, respectively. Furthermore, a query align module along with a scale controller are proposed to enhance the object information around the local queries by limiting the attention to a certain range without losing important information. Finally, the boundaries and types of the objects are acquired from the detection head based on bipartite matching. To verify the effectiveness of the proposed method, comparative experiments were carried out with other state-of-the-art methodologies on two public datasets: the High-Resolution Remote Sensing Detection (HRRSD) dataset and the object detection in optical remote sensing images (DIOR) dataset. The experimental results confirm the effectiveness and superiority of the QETR model, which achieved 71.5% and 91.1% mAP values on the DIOR and HRRSD datasets, respectively.