Change detection based on remote sensing images plays an important role in the field of remote sensing analysis, and it has been widely used in many areas, such as resources monitoring, urban planning, disaster assessment, etc. In recent years, it has aroused widespread interest due to the explosive development of artificial intelligence (AI) technology, and change detection algorithms based on deep learning frameworks have made it possible to detect more delicate changes (such as the alteration of small buildings) with the help of huge amounts of remote sensing data, especially high-resolution (HR) data. Although there are many methods, we still lack a deep review of the recent progress concerning the latest deep learning methods in change detection. To this end, the main purpose of this paper is to provide a review of the available deep learning-based change detection algorithms using HR remote sensing images. The paper first describes the change detection framework and classifies the methods from the perspective of the deep network architectures adopted. Then, we review the latest progress in the application of deep learning in various granularity structures for change detection. Further, the paper provides a summary of HR datasets derived from different sensors, along with information related to change detection, for the potential use of researchers. Simultaneously, representative evaluation metrics for this task are investigated. Finally, a conclusion of the challenges for change detection using HR remote sensing images, which must be dealt with in order to improve the model’s performance, is presented. In addition, we put forward promising directions for future research in this area.

In recent years, building change detection has made remarkable progress through using deep learning. The core problems of this technique are the need for additional data (e.g., Lidar or semantic labels) and the difficulty in extracting sufficient features. In this paper, we propose an end-to-end network, called the pyramid feature-based attention-guided Siamese network (PGA-SiamNet), to solve these problems. The network is trained to capture possible changes using a convolutional neural network in a pyramid. It emphasizes the importance of correlation among the input feature pairs by introducing a global co-attention mechanism. Furthermore, we effectively improved the long-range dependencies of the features by utilizing various attention mechanisms and then aggregating the features of the low-level and co-attention level; this helps to obtain richer object information. Finally, we evaluated our method with a publicly available dataset (WHU) building dataset and a new dataset (EV-CD) building dataset. The experiments demonstrate that the proposed method is effective for building change detection and outperforms the existing state-of-the-art methods on high-resolution remote sensing orthoimages in various metrics.

Carbon sink estimation and ecological assessment of forests require accurate forest type mapping. The traditional survey method is time consuming and labor intensive, and the remote sensing method with high-resolution, multi-spectral commercial satellite images has high cost and low availability. In this study, we explore and evaluate the potential of freely-available multi-source imagery to identify forest types with an object-based random forest algorithm. These datasets included Sentinel-2A (S2), Sentinel-1A (S1) in dual polarization, one-arc-second Shuttle Radar Topographic Mission Digital Elevation (DEM) and multi-temporal Landsat-8 images (L8). We tested seven different sets of explanatory variables for classifying eight forest types in Wuhan, China. The results indicate that single-sensor (S2) or single-day data (L8) cannot obtain satisfactory results; the overall accuracy was 54.31% and 50.00%, respectively. Compared with the classification using only Sentinel-2 data, the overall accuracy increased by approximately 15.23% and 22.51%, respectively, by adding DEM and multi-temporal Landsat-8 imagery. The highest accuracy (82.78%) was achieved with fused imagery, the terrain and multi-temporal data contributing the most to forest type identification. These encouraging results demonstrate that freely-accessible multi-source remotely-sensed data have tremendous potential in forest type identification, which can effectively support monitoring and management of forest ecological resources at regional or global scales.

Airborne laser scanning (ALS) point cloud data are suitable for digital terrain model (DTM) extraction given its high accuracy in elevation. Existing filtering algorithms that eliminate non-ground points mostly depend on terrain feature assumptions or representations; these assumptions result in errors when the scene is complex. This paper proposes a new method for ground point extraction based on deep learning using deep convolutional neural networks (CNN). For every point with spatial context, the neighboring points within a window are extracted and transformed into an image. Then, the classification of a point can be treated as the classification of an image; the point-to-image transformation is carefully crafted by considering the height information in the neighborhood area. After being trained on approximately 17 million labeled ALS points, the deep CNN model can learn how a human operator recognizes a point as a ground point or not. The model performs better than typical existing algorithms in terms of error rate, indicating the significant potential of deep-learning-based methods in feature extraction from a point cloud.

Water-body segmentation in high-resolution satellite imagery is challenging because of the significant variations in the appearance, size, and shape of water bodies. In this letter, a novel multiscale refinement network (MSR-Net) is proposed for water-body segmentation. Similar to most learning-based methods, the MSR-Net resorts to the multiscale information for segmentation, but it improves existing networks in two ways: First, it uses the multiscale information in a new perspective. Instead of the traditional one-off manner that concatenates features and conducts segmentation on one uniform scale, the MSR-Net adopts a new multiscale refinement scheme that makes full use of the multiscale features for more accurate water-body segmentation. In addition, a novel erasing-attention module is designed for an effective feature embedding during the refinement scheme. Experiments on the Gaofen Image Data Set and the DeepGlobe Data Set demonstrate the superiority of MSR-Net when compared with the other state-of-the-art semantic segmentation methods, including U-Net, SegNet, DeepLabv3+, and ExFuse.

Building change detection is useful for land management, disaster assessment, illegal building identification, urban growth monitoring, and geographic information database updating. This study proposes an automatic method that applies object-based analysis to multi-temporal point cloud data to detect building changes. The aim of this building change detection method is to identify areas that have changed and to obtain from-to information. In this method, the data are first preprocessed to generate two sets of digital surface models (DSMs), digital elevation models, and normalized DSMs from registered old and new point cloud data. Thereafter, on the basis of differential DSM, candidates for changed building objects are identified from the points in the smooth areas by using a connected component analysis technique. The random sample consensus fitting algorithm is then used to distinguish the true changed buildings from trees. The changed building objects are classified as “newly built”, “taller”, “demolished” or “lower” by using rule-based analysis. Finally, a test data set consisting of many buildings of different types in an 8.5 km² area is selected for the experiment. In the test data set, the method correctly detects 97.8% of buildings larger than 50 m². The accuracy of the method is 91.2%. Furthermore, to decrease the workload of subsequent manual checking of the result, the confidence index for each changed object is computed on the basis of object features.