Search

ABSTRACTAccurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Our method involves three steps: (1) Image encoder and edge node initialization, (2) Image feature fusion with an enhanced segmentation refinement branch, and (3) Edge filtering and structural reasoning. Our method outperforms previous works on the vectorizing world building dataset and the Enschede dataset, with vertex and edge heat map F1-scores increasing from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], and from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], respectively. Furthermore, our method demonstrates superior performance compared to the current state-of-the-art based on qualitative evaluations, indicating its effectiveness in extracting global image information while maintaining the consistency and topological validity of the roof structure.

Current land use classification models based on very high-resolution (VHR) remote sensing images often suffer from high sample dependence and poor transferability. To address these challenges, we propose an unsupervised multisource domain adaptation framework for cross-domain land use classification that eliminates the need for repeatedly using source domain data. Our method uses the Swin Transformer as the backbone of the source domain model to extract features from multiple source domain samples. The model is trained on source domain samples, and unlabeled target domain samples are then used for target domain model training. To minimize the feature discrepancies between the source and target domains, we use a weighted information maximization loss and self-supervised pseudolabels to alleviate cross-domain classification noise. We conducted experiments on four public scene datasets and four new land use scene datasets created from different VHR images in four Chinese cities. Results show that our method outperformed three existing single-source cross-domain methods (i.e., DANN, DeepCORAL, and DSAN) and four multisource cross-domain methods (i.e., M3SDA, PTMDA, MFSAN, and SHOT), achieving the highest average classification accuracy and strong stability. We conclude that our method has high potential for practical applications in cross-domain land use classification using VHR images.

Height map estimation from a single aerial image plays a crucial role in localization, mapping, and 3D object detection. Deep convolutional neural networks have been used to predict height information from single-view remote sensing images, but these methods rely on large volumes of training data and often overlook geometric features present in orthographic images. To address these issues, this study proposes a gradient-based self-supervised learning network with momentum contrastive loss to extract geometric information from non-labeled images in the pretraining stage. Additionally, novel local implicit constraint layers are used at multiple decoding stages in the proposed supervised network to refine high-resolution features in height estimation. The structural-aware loss is also applied to improve the robustness of the network to positional shift and minor structural changes along the boundary area. Experimental evaluation on the ISPRS benchmark datasets shows that the proposed method outperforms other baseline networks, with minimum MAE and RMSE of 0.116 and 0.289 for the Vaihingen dataset and 0.077 and 0.481 for the Potsdam dataset, respectively. The proposed method also shows around threefold data efficiency improvements on the Potsdam dataset and domain generalization on the Enschede datasets. These results demonstrate the effectiveness of the proposed method in height map estimation from single-view remote sensing images.

Objectives: In vivo magnetic resonance spectroscopy (MRS) was used to investigate neurometabolic homeostasis in children with functional neurological disorder (FND) in three regions of interest: supplementary motor area (SMA), anterior default mode network (aDMN), and posterior default mode network (dDMN). Metabolites assessed included N-acetyl aspartate (NAA), a marker of neuron function; myo-inositol (mI), a glial-cell marker; choline (Cho), a membrane marker; glutamate plus glutamine (Glx), a marker of excitatory neurotransmission; γ-aminobutyric acid (GABA), a marker of inhibitor neurotransmission; and creatine (Cr), an energy marker. The relationship between excitatory (glutamate and glutamine) and inhibitory (GABA) neurotransmitter (E/I) balance was also examined. Methods: MRS data were acquired for 32 children with mixed FND (25 girls, 7 boys, aged 10.00 to 16.08 years) and 41 healthy controls of similar age using both short echo point-resolved spectroscopy (PRESS) and Mescher-Garwood point-resolved spectroscopy (MEGAPRESS) sequences in the three regions of interest. Results: In the SMA, children with FND had lower NAA/Cr, mI/Cr (trend level), and GABA/Cr ratios. In the aDMN, no group differences in metabolite ratios were found. In the pDMN, children with FND had lower NAA/Cr and mI/Cr (trend level) ratios. While no group differences in E/I balance were found (FND vs. controls), E/I balance in the aDMN was lower in children with functional seizures—a subgroup within the FND group. Pearson correlations found that increased arousal (indexed by higher heart rate) was associated with lower mI/Cr in the SMA and pDMN. Conclusions: Our findings of multiple differences in neurometabolites in children with FND suggest dysfunction on multiple levels of the biological system: the neuron (lower NAA), the glial cell (lower mI), and inhibitory neurotransmission (lower GABA), as well as dysfunction in energy regulation in the subgroup with functional seizures.

This study focuses on the development of a new framework for evaluating bikeability in urban environments with the aim of enhancing sustainable urban transportation planning. To close the research gap that previous studies have disregarded the dynamic environmental factors and trajectory data, we propose a framework that comprises four sub-indices: safety, comfort, accessibility, and vitality. Utilizing open-source data, advanced deep neural networks, and GIS spatial analysis, the framework eliminates subjective evaluations and is more efficient and comprehensive than prior methods. The experimental results on Xiamen, China, demonstrate the effectiveness of the framework in identifying areas for improvement and enhancing cycling mobility. The proposed framework provides a structured approach for evaluating bikeability in different geographical contexts, making reproducing bikeability indices easier and more comprehensive to policymakers, transportation planners, and environmental decision-makers.

Stress exposures and dysregulated responses to stress are implicated in psychiatric disorders of mood, anxiety, and cognition. Perceived stress, an individual's appraisal of experienced stress and ability for coping, relates to dysregulated functioning in resting state brain networks. Alterations in GABAergic function may underlie perceived stress-related functional dysregulation in resting state networks but this has not yet been explored. Therefore, the current study examined the association of perceived stress, via the Perceived Stress Scale (PSS), with prefrontal GABA levels and corresponding resting state functional connectivity (RSFC) alterations. Twelve women and five men, ages 35–61, participated. MR spectroscopy was used to measure brain GABA levels in the anterior cingulate cortex (ACC), left dorsolateral prefrontal cortex (DLPFC), and ventromedial prefrontal cortex (VMPFC). Resting state functional scans acquired at 3 Tesla were used to measure RSFC within and between the default mode (DMN), salience (SN), and central executive networks (CEN), hippocampus and amygdala. We observed significant negative correlations between total PSS scores and left DLPFC GABA levels (r = −0.62, p = 0.023). However, PSS scores were not significantly correlated with RSFC measures (all p > 0.148). These preliminary results support a relationship between perceived stress and GABAergic functioning in DLPFC, a core node of the CEN, an intrinsic network thought to underlie goal-directed attentional processes. Our findings extend previous work suggesting that functioning in the CEN is related to perceived stress and may inform treatment strategies to improve outcomes in stress-related conditions.

The automatic identification of urban functional regions (UFRs) is crucial for urban planning and management. A key issue involved in URF classification is to properly determine the basic functional units, for which popular practices are usually based upon existing land use boundaries or road networks. Such practices suffer from the unavailability of existing datasets, leading to difficulty in large-scale mapping. To deal with this problem, this paper presents a method to automatically obtain functional units for URF classification using high-resolution remote sensing images. We develop a context-aware segmentation network to simultaneously extract buildings and road networks from remote sensing images. The extracted road networks are used for partitioning functional units, upon which five main building types are distinguished considering building height, morphology, and geometry. Finally, the UFRs are classified according to the distribution of building types. We conducted experiments using a GaoFen-2 satellite image with a spatial resolution of 0.8 m acquired in Fuzhou, China. Experimental results showed that the proposed segmentation network performed better than other convolutional neural network segmentation methods (i.e., PSPNet, Deeplabv3+, DANet, and JointNet), with an increase of F1-score up to 1.37% and 1.19% for road and building extraction, respectively. Results also showed that the residential regions, accounting for most of the urban areas, identified by the proposed method had a user accuracy of 94%, implying the promise of the proposed method for deriving the spatial units and the types of urban functional regions.

Deep learning-based models for building delineation from remotely sensed images face the challenge of producing precise and regular building outlines. This study investigates the combination of normalized digital surface models (nDSMs) with aerial images to optimize the extraction of building polygons using the frame field learning method. Results are evaluated at pixel, object, and polygon levels. In addition, an analysis is performed to assess the statistical deviations in the number of vertices of building polygons compared with the reference. The comparison of the number of vertices focuses on finding the output polygons that are the easiest to edit by human analysts in operational applications. It can serve as guidance to reduce the post-processing workload for obtaining high-accuracy building footprints. Experiments conducted in Enschede, the Netherlands, demonstrate that by introducing nDSM, the method could reduce the number of false positives and prevent missing the real buildings on the ground. The positional accuracy and shape similarity was improved, resulting in better-aligned building polygons. The method achieved a mean intersection over union (IoU) of 0.80 with the fused data (RGB + nDSM) against an IoU of 0.57 with the baseline (using RGB only) in the same area. A qualitative analysis of the results shows that the investigated model predicts more precise and regular polygons for large and complex structures.

Accurate boundaries of smallholder farm fields are important and indispensable geo-information that benefits farmers, managers, and policymakers in terms of better managing and utilizing their agricultural resources. Due to their small size, irregular shape, and the use of mixed-cropping techniques, the farm fields of smallholder can be difficult to delineate automatically. In recent years, numerous studies on field contour extraction using a deep Convolutional Neural Network (CNN) have been proposed. However, there is a relative shortage of labeled data for filed boundaries, thus affecting the training effect of CNN. Traditional methods mostly use image flipping, and random rotation for data augmentation. In this paper, we propose to apply Generative Adversarial Network (GAN) for the data augmentation of farm fields label to increase the diversity of samples. Specifically, we propose an automated method featured by Fully Convolutional Neural networks (FCN) in combination with GAN to improve the delineation accuracy of smallholder farms from Very High Resolution (VHR) images. We first investigate four State-Of-The-Art (SOTA) FCN architectures, i.e., U-Net, PSPNet, SegNet and OCRNet, to find the optimal architecture in the contour detection task of smallholder farm fields. Second, we apply the identified optimal FCN architecture in combination with Contour GAN and pixel2pixel GAN to improve the accuracy of contour detection. We test our method on the study area in the Sudano-Sahelian savanna region of northern Nigeria. The best combination achieved F1 scores of 0.686 on Test Set 1 (TS1), 0.684 on Test Set 2 (TS2), and 0.691 on Test Set 3 (TS3). Results indicate that our architecture adapts to a variety of advanced networks and proves its effectiveness in this task. The conceptual, theoretical, and experimental knowledge from this study is expected to seed many GAN-based farm delineation methods in the future. [ABSTRACT FROM AUTHOR]