Inner-city redevelopment is regarded as an effective way to promote land-use efficiency and optimize land-use structure, especially with the establishment of urban growth boundaries in Chinese cities. However, inner-city redevelopment as compared to urban sprawl has been rarely monitored in 2D space, let alone in 3D space. Therefore, in this paper, a novel approach to generate time-series 3D building maps (i.e., building footprint and height) based on high-resolution (2 m) multi-view ZY-3 satellite imagery was proposed. In the proposed method, the building footprint was updated by an object-based image-to-map change detection method, which employed spectral (i.e., HSV and NDVI) and structural features (i.e., morphological building index) to extract non-building and building objects, respectively; building height was estimated automatically through semi-global matching of multi-view images. We applied the proposed method to four representative Chinese megacities, i.e., Beijing, Xi’an, Shanghai, and Wuhan, for the period 2012–2017, and detected building footprints with overall accuracies ranging from 84.84% to 97.60%. The building height estimation was also relatively accurate, with the bias, slope, and root-mean-square error being −0.49–2.30 m, 0.93–1.10 m, and 4.94–7.31 m, respectively. Our results show that the total building coverage decreased over the study period, accompanied by an increase in both area-weighted building height and floor area ratio. In addition, compact low-rise buildings have been replaced by open high-rise buildings in the urban redevelopment process. Moreover, due to the scattered spatial distribution of the redevelopment sites, the local spatial aggregation patterns of building density are unlikely to shift between hotspots (i.e., spatial aggregation of high values) and coldspots (i.e., spatial aggregation of low values).

Change detection of remote sensing images has a wide range of applications in many fields. In recent years, deep learning has become one of the most powerful tools for remote sensing change detection, thanks to its excellent feature learning capability. However, most deep learning methods require a lot of labeled data for the training, which is time-consuming and labor-intensive. Recently, a new learning paradigm—self-supervised learning—has become one of the hot topics in the field of change detection due to its ability to learn feature representations by training with a large amount of unlabeled data and without a large number of sample annotations. However, the existing methods for self-supervised learning are usually designed for natural image processing and are less considered for change detection in more complex scenes (e.g., remote sensing imagery). Therefore, in this paper, we propose a novel network by fusing temporal and differential information for self-supervised contrastive learning change detection, namely TD-SSCD. Specifically, TD-SSCD aims to mine information from the bi-temporal images and their differential images in a self-supervised learning framework, and it gradually learns the potential correlations between them through an alternating iteration learning strategy. The experimental results based on the OSCD and SZTAKI datasets show that the proposed method outperforms the current state-of-the-art unsupervised and self-supervised change detection methods. Benefiting from pre-training on unlabeled samples, the method closes the gap between unsupervised and supervised change detection.

Multi-view stereo (MVS) aerial image depth estimation is a research frontier in the remote sensing field. Recent deep learning based advances in close-range object reconstruction have suggested the great potential of this approach. Meanwhile, the deformation problem and the scale variation issue are also worthy of attention. These characteristics of aerial images limit the applicability of the current methods for aerial image depth estimation. Moreover, there are few available benchmark datasets for aerial image depth estimation. In this regard, this paper describes a new benchmark dataset called the LuoJia-MVS dataset, as well as a new deep neural network known as HDC-MVSNet. The LuoJia-MVS dataset contains 7972 five-view images with a spatial resolution of 10 cm, pixel-wise depths, and precise camera parameters, and was generated from an accurate digital surface model built from thousands of stereo aerial images. In the HDC-MVSNet network, a new full-scale feature pyramid extraction module, a hierarchical set of three-dimensional (3D) convolutional blocks, and “true 3D” deformable 3D convolutional layers are specifically designed by considering the afore-mentioned characteristics of aerial images. Overall and ablation experiments on the WHU and LuoJia-MVS datasets validated the superiority of HDC-MVSNet over the current state-of-the-art MVS depth estimation methods, and confirmed that the newly built dataset can provide an effective benchmark.

In recent years, numerous change detection methodologies have been proposed, with a predominant focus on binary change detection. Furthermore, there exists a paucity of research addressing semantic change detection in scenarios where solely binary change labels are available. This article introduces a multitask network for semantic change detection. First, 3-D ResUnet model is employed to generate initial multitemporal land cover results through postclassification comparison. Subsequently, the multitask network, encompassing two subtasks—binary change detection and multitemporal semantic segmentation—is proposed. Specifically, the shared branch of the network employs 3-D residual blocks to extract joint spectral-spatial features. In the subsequent task-specific branch, a 3-D GAN is incorporated for the binary change detection task to enhance the discrimination ability of latent high-level features for changes. Novel adaptive self-paced learning and certainty-weighted focal loss are proposed for multitemporal semantic segmentation to mitigate adverse effects from noisy semantic labels by considering sample complexity and reliability in the network optimization process. Experiments conducted on the Orbita Hyperspectral dataset in the Xiong'an New Area demonstrate the superior performance of the proposed method, achieving 99.28% and 76.60% for overall accuracy and kappa, respectively. This outperformance is notable when compared to other methods, such as Str4 and Bi-SRNet, showing an increase of 39.82% and 54.17% for kappa. Moreover, comparative experiments on SECOND further confirm the advantage of the proposed method, achieving 54.62% for kappa and outperforming other comparative methods, such as Bi-SRNet (47.61%).

Assessment of ecosystem services supply, demand, and budgets can help to achieve sustainable urban development. The Guangdong-Hong Kong-Macao Greater Bay Area, as one of the most developed megacities in China, sets up a goal of high-quality development while fostering ecosystem services. Therefore, assessing the ecosystem services in this study area is very important to guide further development. However, the spatial pattern of ecosystem services, especially at local scales, is not well understood. Using the available 2017 land cover product, Sentinel-1 SAR and Sentinel-2 optical images, a deep learning land cover mapping framework integrating deep change vector analysis and the ResUnet model was proposed. Based on the produced 10 m land cover map for the year 2020, recent spatial patterns of the ecosystem services at different scales (i.e., the GBA, 11 cities, urban–rural gradient, and pixel) were analyzed. The results showed that: (1) Forest was the primary land cover in Guangzhou, Huizhou, Shenzhen, Zhuhai, Jiangmen, Zhaoqing, and Hong Kong, and an impervious surface was the main land cover in the other four cities. (2) Although ecosystem services in the GBA were sufficient to meet their demand, there was undersupply for all the three general services in Macao and for the provision services in Zhongshan, Dongguan, Shenzhen, and Foshan. (3) Along the urban–rural gradient in the GBA, supply and demand capacity showed an increasing and decreasing trend, respectively. As for the city-level analysis, Huizhou and Zhuhai showed a fluctuation pattern while Jiangmen, Zhaoqing, and Hong Kong presented a decreasing pattern along the gradient. (4) Inclusion of neighborhood landscape led to increased demand scores in a small proportion of impervious areas and oversupply for a very large percent of bare land.

The land use (LU) classification of remote sensing (RS) images has broad applications in various fields. In recent years, hybrid convolutional neural networks (CNNs)–Transformer models have been widely applied to the LU classification task. These methods combine the advantages of CNNs and transformers to mine information from RS images fully for LU classification. However, due to rich content and complex backgrounds in RS images, existing methods often overlook the deep semantic information and spatial structural relationships hidden within the optical domain, resulting in limited classification accuracy. To address these challenges, we propose a novel LU classification model: the global–local interactive feature fusion model (GLIFFM). The GLIFFM utilizes a CNN–Transformer architecture and simultaneously integrates land cover (LC) primitives to fully exploit both semantic content within the RS images and the spatial structural information. First, a mutual feature extraction (MFE) module is designed to extract correlated and complementary features between optical and LC primitives. Next, we introduce a local and global feature enhancement module (LGFEM) to enhance deep features containing both LC and optical information. Finally, a gated information fusion block (GIFB) is proposed to perform gated fusion of the local and global features. The experimental results conducted on the WHDLD, EuroSAT, and LCZ-Guangdong-Hong Kong-Macao Greater Bay Area (LCZGBA) datasets demonstrate that our GLIFFM can achieve superior classification performance and outperform many state-of-the-art methods. Our source codes are available at https://github.com/Fimwu7/The-GLIFFM-for-Land-Use-Classifcation.