Extracting buildings from high-resolution remote sensing images is essential for many geospatial applications, such as building change detection, urban planning, and disaster emergency assessment. Due to the diversity of geometric shapes and the blurring of boundaries among buildings, it is still a challenging task to accurately generate building footprints from the complex scenes of remote sensing images. The rapid development of convolutional neural networks is presenting both new opportunities and challenges with respect to the extraction of buildings from high-resolution remote sensing images. To capture multilevel contextual information, most deep learning methods extract buildings by integrating multilevel features. However, the differential responses between such multilevel features are often ignored, leading to blurred contours in the extraction results. In this study, we propose an end-to-end multitask building extraction method to address these issues; this approach utilizes the rich contextual features of remote sensing images to assist with building segmentation while ensuring that the shape of the extraction results is preserved. By combining boundary classification and boundary distance regression, clear contour and distance transformation maps are generated to further improve the accuracy of building extraction. Subsequently, multiple refinement modules are used to refine each part of the network to minimize the loss of image feature information. Experimental comparisons conducted on the SpaceNet and Massachusetts building datasets show that the proposed method outperforms other deep learning methods in terms of building extraction results.

Accurate and complete extraction of buildings from very high resolution (VHR) remote sensing (RS) images is highly important for urban planning and land management. However, owing to the limited information available for small buildings and building boundaries, as well as challenges such as the spectral similarity of ground objects, tree occlusions, and shadow interference, automatically extracting buildings from VHR images remains challenging. These issues may result in building extraction errors such as misclassification, small building omission, blurred boundaries, and incorrect segmentation. To address these challenges, we propose a multiscale hybrid transformer (MSHFormer) with boundary enhancement. This approach incorporates a hybrid encoder that combines a multiscale local perception module (MSLP) and a global perception module (GPM), combining the strengths of convolutional neural networks (CNNs) and transformers to achieve efficient synergy between global modelling and local feature extraction. Additionally, we developed an edge enhancement module (EHM) to enhance boundary information, significantly improving building boundary segmentation accuracy. Finally, we design a group alignment feature fusion module (GAFFM) to efficiently integrate low-level features from the encoder with high-level features from the decoder, reducing feature space misalignment. The experimental results on three public datasets demonstrate the effectiveness of MSHFormer. Specifically, the proposed method achieves intersection-over-union (IoU) values of 89.1%, 73.6%, and 89.5% on the Potsdam, Massachusetts, and WHU datasets, respectively.

Traditional deep semantic segmentation methods rely heavily on large-scale, densely labeled data, which is costly and time-consuming to obtain. Semi-supervised frameworks have emerged as a promising alternative, reducing the dependency on pixel-level annotations while enhancing segmentation performance. However, these methods still face challenges in data augmentation and class imbalance. For instance, traditional augmentation techniques such as CutMix are limited by their reliance on single random local contexts, which restricts perturbation strength and hinders model generalization. Moreover, the long-tailed class distribution in remote sensing (RS) images is often overlooked, leading to pseudo-labels that are biased toward majority classes, further exacerbating class imbalance. To address these challenges,we propose a novel semi-supervised framework for RS semantic segmentation, named APRW, which incorporates adaptive patch augmentation and class rank weighting. The adaptive patch augmentation module dynamically generates adaptive patch masks by comparing the predictions of weakly and strongly augmented inputs. These masks focus on high-confidence regions while preserving diverse boundary structures, effectively expanding the perturbation space and enhancing the model’s adaptability to varied inputs. The class rank weighting module maintains memory banks for both labeled and unlabeled data, dynamically calculates class weights based on the relative ranking of pixel confidences, and adaptively fuses these weights. This design mitigates biases caused by long-tailed distributions, improves the segmentation of rare classes, and enhances overall segmentation accuracy. Extensive experiments demonstrate that our method consistently outperforms existing approaches across multiple RS datasets, including DFC22, iSAID, MER, MSL, Vaihingen, and GID-15, showcasing superior generalization and higher segmentation precision. Implementation is released at https://github.com/135az/APRW.

As urbanization accelerates, more and taller buildings and less greenery are closely related to changes in the urban thermal environment (UTE). Knowledge of spatial and temporal variations of UTE is becoming increasingly concerning, and this can be measured with the land surface temperature (LST). Satellite observation of LST is an important tool for monitoring; the strong thermal anisotropy limits the use of satellite thermal infrared (TIR) data. Hitherto, the poor investigation was focused on the modeling and analysis of urban thermal anisotropy (UTA), especially in mountainous urban areas with multi-slope environments. These areas exhibit a distinctive “roof effect”, which is defined as the radiative transfer effect between the roof and the adjacent wall due to the slope that results in different heights between the roofs; multiple scattering has also been changed. Although an analytical thermal anisotropy model for the urban canopy over sloping terrain (AU3SM) has been proposed, its inability to effectively account for roof effects and multi-scattering mechanisms limits its daytime TIR observation applicability. To address these limitations, we developed an enhanced AU3SM that considers the roof effect and multiple scattering, which is labeled AU3SM-RS. The model was evaluated using measurements based on unmanned aerial vehicles (UAVs) in the mountainous city of Chongqing, China, with values of the root mean square error (RMSE) and coefficient of determination (R²) of 0.83 K and 0.93 in UTA. Comparison with a graphic processing unit-based solution for the faster 3-D radiative transfer model (GRay) further validates the model’s reliability with RMSE and R² values of 0.12 K and 0.96, respectively. Simulations in a certain scenario reveal that as the slope increases, the roof effect increases and the multiple scattering effect decreases in UTA and brightness temperature (BT), ignoring the roof effect and multiple scattering can result in maximum UTA biases of approximately 0.54, 0.48, and 0.72 K, BT biases approximately 1.02, 1.62, and 2.4 K at 5°, 15°, and 30° slopes, the biases due to neglecting the second scattering are very slight compared to the first scattering. Under certain conditions with a slope of 10°, the wider roof and narrower roadway, a related more dramatic roof effect; the narrower and deeper street canyon, a related more dramatic scattering effect. The proposed model is an efficient computational tool to assess UTA in mountainous areas quickly.