Search Results (125)

Deep neural networks increasingly convert street-level imagery into quantitative measures of urban perception, but the cost of transformer backbones limits repeated inference over city-scale image collections. This study proposes an optimized Sparse Attention Regularized Transformer with a Low-Rank constraint (SART-LR), a Siamese vision transformer in which softmax is replaced by learnable $\alpha$-entmax attention and the attention and feed-forward projections are directly factorized. The evaluation assigns each Place Pulse 2.0 image to exactly one of the training, validation, or test partitions, thereby preventing the same image from entering multiple partitions through different comparisons. All models are tuned with an equal, architecture-specific validation budget and evaluated over five seeds. Under this image-disjoint evaluation, SART-LR reaches an average pairwise accuracy of 72.6%, 2.1 percentage points above ViT-B/16 and 1.2 points above Swin-T. The explicit layer-wise derivation gives 2.97 million trainable parameters, a 5.0-fold reduction relative to the depth- and width-matched dense backbone and a 29.1-fold difference from ViT-B/16; the latter comparison is reported only as an end-to-end model total because the architectures differ. Five repeated city-level folds give a cross-city accuracy of $67.0\pm 1.3$%, corresponding to a 5.6-point decrease from the image-disjoint result. Paired tests indicate that the advantage over Swin-T varies by attribute, and analyses stratified by rater agreement show lower accuracy for ambiguous comparisons. These findings support an accuracy–efficiency benefit on Place Pulse 2.0, while the absence of an independent urban-perception dataset and the geographic imbalance of the 56-city sample limit external-validity claims. Full article

High-precision urban vegetation mapping is essential for assessing carbon sink capacities, mitigating the urban heat island effect, and supporting sustainable development. Although deep learning and high-resolution remote sensing have advanced automated vegetation monitoring, existing models still face challenges when a common segmentation architecture is evaluated under different imaging geometries. In this study, Cityscapes and ISPRS Vaihingen are treated as two independent benchmarks representing perspective street-level imagery and orthographic aerial imagery, rather than as simultaneous cross-view inputs. “Background dominance” caused by perspective distortion and the “gridding artifacts” inherent in orthographic textures severely constrain segmentation accuracy across varying vegetation scales, particularly for small targets. To address these limitations, we propose a Scale-Aware Mixture of Experts (SA-MoE) architecture for fine-grained vegetation segmentation under two distinct imaging views, together with a scene-specific optimization strategy. The core SA-MoE framework consists of two main components. First, the spatial gating network uses a temperature polarization mechanism with τ = 0.5 to adjust the initial logit maps, sharpening expert-weight differences while preserving stable gradient propagation. Second, we use a heterogeneous expert group with five parallel branches: a pixel-level expert, three spatial experts with different dilation rates, and a global average-pooling expert. A dynamic pixel-level weighted fusion mechanism is then applied, decoupling feature extraction from receptive-field allocation. Furthermore, to address the heterogeneity of “hard samples” and “label noise” across the two benchmark settings, we introduce a scene-specific optimization strategy. Our findings show that the Focal-Dice (FD) loss is more suitable for perspective scenes with severe target imbalance and hard-to-classify vegetation targets, whereas the Cross-Entropy (CE) loss is more robust to boundary jitter in orthographic imagery. Comparative experiments on the Cityscapes (perspective view) and ISPRS Vaihingen (orthographic view) datasets reveal that SA-MoE achieves a highly competitive balance between computational efficiency and fine-grained segmentation, particularly in micro-target recall. Notably, the recall for extra-small (XS) scale targets in the aerial dataset improved by 3.21 percentage points compared to the second-best model. For the street-level dataset, our model achieved competitive global performance in terms of Overall Accuracy (OA), Precision, and F1-Score. However, we also observed a performance trade-off, where Transformer-based models maintained an advantage in preserving fine boundary details for these extra-small targets. In the routing analysis, we observed a pattern that we refer to as “receptive field inversion”, in which the model assigns lower weights to large-dilation experts for large canopy regions in orthophotos. We interpret this pattern as a plausible routing hypothesis. Overall, SA-MoE offers an efficient and adaptive solution for urban vegetation mapping under two imaging views. Full article

How the diffusion of green innovation technologies translates into regional landscape sustainability is still poorly resolved, in part because most studies rely on a single data source that cannot separate an adoption signal from confounding climatic and terrain influences. To make progress on this identification problem, an empirical framework is assembled that fuses openly licensed observations, Landsat and Sentinel-2 imagery, OpenStreetMap layers, NPP-VIIRS nighttime lights, and public statistical yearbooks, and embeds them in a spatial econometric design, so that the adoption–pattern–service–sustainability chain can be traced across 72 county-level units spanning Ningxia, eastern Gansu, and northern Shaanxi over 2013–2022. Pixel- and object-level integration, entropy weighting, and principal component reduction jointly deliver a fused representation whose coefficient of determination against held-out reference data exceeds 0.85 while the reconstruction error falls by roughly a third relative to single-source baselines. A spatial Durbin specification then decomposes adoption’s association with sustainability into a dominant direct component and a smaller, distance-bounded spillover, and roughly one-quarter of the total travels through landscape reconfiguration; the result survives the placebo, subsample, and variable-substitution checks, and is strongly conditioned by the terrain and local economic capacity. These findings favour a spatially coordinated, capacity-targeted transition policy rather than uniform deployment. Two caveats should be read alongside them: adoption is measured through proxies whose validity, though corroborated against county-level green-patent and installed-capacity records, is not perfect, and external validation across contrasting landscapes remains outstanding. Full article

Accurate greenspace quantification underpins sustainable urban greening management. Remote sensing (RS)-derived green quantity is a core urban planning indicator, yet its capability to reflect actual urban green supply lacks systematic verification, inevitably affecting planning formulation and decision-making. Taking Shanghai as the study area, this study integrates remote sensing and Baidu Street View (BSV data) to compare the two-dimensional planar green coverage derived from satellite imagery with pedestrian-level perceived visual green coverage. The Mask2Former model was adopted for high-precision semantic segmentation of BSV images to extract vegetation, building, sky and hard pavement proportions. Geographically Weighted Regression (GWR), hotspot analysis and transition mapping were applied to identify divergent regions, while the XGBoost-SHAP framework was employed to explore deviation mechanisms. The results reveal distinct spatial pattern differences between the two green quantity datasets. RS-derived green quantity exhibits strip-like agglomeration, whereas BSV-perceived green quantity is more fragmented, with a correlation coefficient of only 0.240. Single RS quantification fails to reflect street-level green supply. Divergent areas are classified into accurate, overestimated and underestimated zones. RS underestimates green quantity in central urban areas and overestimates that in northwest suburbs. BSV-based sky ratio, building ratio, hardscape ratio and Road 1 density are core influencing factors with obvious nonlinear threshold effects. This study clarifies the quantitative deviation patterns and mechanisms between RS and BSV green quantity, providing scientific support for precise urban green planning and sustainable perceived visual green construction. Practically, the dual RS–street view assessment framework proposed in this paper can be embedded into routine urban green infrastructure auditing, help planners distinguish systematically overestimated suburban green belts and underestimated central urban micro-green spaces, and deliver targeted optimization strategies for vertical greening, street tree renovation and pocket park construction under high-quality urban renewal demands. Full article

Rapid urbanization has intensified inequalities in the built environment (BE), yet their relationship with well-being in areas where vulnerable populations are concentrated remains unclear. Using Shanghai as a case, this study examines the association between BE equity and area-level happiness across five types of vulnerable areas: those with larger numbers of minors, older adults, or migrant residents; economically disadvantaged areas; and areas with limited educational resources. We integrated geotagged Weibo check-in texts collected from January to December 2023, multi-source spatial data, street-view imagery, and interpretable machine learning. ERNIE 3.0 was used to derive sentiment scores from individual posts, which were then aggregated within each spatial unit to construct an area-level happiness score. Thirty-four BE indicators were organized within the 5D framework and analyzed using Mask2Former, Moran’s I, Gini coefficients, LightGBM, and SHAP. Results reveal marked central–peripheral disparities in BE resources and happiness, together with nonlinear and group-specific spatial associations. High-Gini patterns are evident in peri-urban areas, accounting for approximately 67% of vulnerability-focused areas, while strict equity–happiness discordance accounts for 42%. These findings suggest that BE distributional imbalance, adverse high-inequality–low-happiness co-occurrence, and equity–happiness discordance should be distinguished for targeted and inclusive planning. Full article

Growing evidence suggests that urban greenery is associated with improved mental health, yet how eye-level exposure functions within specific socio-environmental contexts remains underexplored. This study presents an exploratory case study utilizing semantic segmentation of street-view imagery to quantify eye-level greenery (“open greenery”) across two contrasting community contexts: a densely developed area (Region 1) and a less developed area (Region 2) in Korea. Using interaction models reinforced by 5000-iteration bootstrap analyses, we identified the moderating role of greenery in the relationship between mental health (depression and cognitive function) and life satisfaction. Our findings indicate that the psychological benefits of greenery are highly contingent upon the interplay between individual vulnerability and regional context. Specifically, in Region 1, greenery moderated well-being for the low-cognitive function subgroup, while in Region 2, the moderating effect was most pronounced among individuals with depressive symptoms. Despite the inherent limitations of small subgroup samples, the stability of these patterns across repeated bootstrap iterations highlights meaningful “spatial intersections” where greenery plays a role in shaping psychological well-being. By adopting a case-centric approach, this study highlights that the benefits of street-view greenery are not uniform but context-dependent. These results underscore the necessity of context-aware green infrastructure strategies tailored to the specific environmental needs of vulnerable populations in diverse community settings. Full article

Urban systems exhibit multifractal scaling behaviors arising from their hierarchical and heterogeneous spatial organization. Yet the Green View Index (GVI), a widely used indicator of street-level green visibility, is conventionally measured as a static value, overlooking how green exposure is reorganized across scales, which may bias greening evaluations and planning decisions. This study proposes an “X-minute Greenery” framework to examine GVI as a scale-dependent spatial process. Street view imagery and walking isochrone data from Hong Kong and Shenzhen, China, are integrated to measure GVI across 5, 10, 15, and 20 min walking ranges, while a multifractal-inspired approach is developed to characterize its scale-dependent variation structure. Empirical findings show that (1) GVI variation is not random but associated with urban governance and spatial morphology; (2) a staged pattern of variation is identified, with a GVI level around 0.2 marking an approximate shift in configurations of greening dynamics; (3) even under similar initial GVI conditions, areas can still follow divergent variation patterns across walking scales, exposing latent spatial inequalities aligned with institutional and morphological factors. By framing urban green visibility as a scale-aware complex-systems phenomenon, this study adapts multifractal reasoning to street-level greenery, challenges the “equal green, equal policies” assumption, and supports context-sensitive planning interventions. Full article

Urban flood detection requires visual examples of flooded streets and alleys, but such event-state images are difficult to collect at scale. This study examines whether sparse real-image training sets can be strengthened using text-prompt-only AI-generated virtual imagery for ground-level flood detection in flood-prone urban areas. Building on AlleyFloodNet, the generated images were used only as condition-specific training augmentation, while validation and testing were conducted exclusively on real images. Across eight ImageNet-pretrained architectures and three random seeds, the results show that virtual imagery is not a substitute for real flood observations. When virtual images dominated the training set, performance declined. However, when a sufficient real-image anchor was available, virtual augmentation improved the highest observed fixed-test performance. The strongest mixed-condition result was obtained by EfficientNet-B2 under Real30_Aug70, reaching 91.00% accuracy, 89.48% flooded-class recall, and 90.79% macro-F1. These findings suggest that prompt-only virtual imagery can help mitigate real-image scarcity in selected training conditions, but its benefit depends on the real-to-virtual composition and model architecture. Full article

Urban heat island effects increasingly challenge pedestrian mobility by intensifying thermal stress and reducing the attractiveness of walking during hot periods. However, most pedestrian routing systems still prioritize distance or travel time, while environmental conditions such as heat exposure and shade are rarely incorporated into operational route generation. Existing comfort-aware approaches often rely on static maps, simulated microclimatic indicators, or descriptive greenery measures, limiting their direct integration into user-configurable pedestrian navigation. This study develops a thermal comfort-aware pedestrian routing framework that integrates heterogenic data sources including observed land surface temperature, pedestrian-perspective tree-canopy coverage, and network distance into a unified multi-criteria pathfinding model. The workflow proceeds in four steps: first, airborne thermal imagery is processed to derive a high-resolution land surface temperature layer; second, Google Street View images are sampled at street-segment locations and segmented using SegFormer to extract visible tree-canopy coverage; third, both environmental indicators are aggregated to a cleaned pedestrian network; and fourth, normalized distance, temperature, and canopy attributes are combined through a user-adjustable edge-cost formulation and solved using Dijkstra’s algorithm. The framework is implemented as an operational web-based routing tool for the historic center of Clermont-Ferrand, France. The routable graph includes 551 nodes and 796 edges, with 600 segments carrying GSV-derived canopy information and 623 segments carrying airborne-derived LST values. Across the network, we observed LST ranges from 19.5 °C to 39.1 °C, while canopy coverage ranged from 0 to 70.6%. For a representative origin–destination pair, the coolest route reduces average LST by nearly 5 °C and almost triples canopy coverage compared with the shortest path, although at the cost of a 72% longer distance. These results demonstrate that the framework can generate interpretable comfort–efficiency trade-offs and support user-comfort pathfinding as an operational approach for heat-resilient pedestrian navigation. Full article

Monitoring the vertical progress of reinforced-concrete buildings supports construction management, urban analytics, and seismic exposure classification, yet camera-based floor counting faces two obstacles: public datasets depict almost exclusively completed structures, and the number of structurally finished floors is visually ambiguous while a building is still being erected. We reformulate building-height estimation as discrete floor-count classification from a single street-level facade image and assemble a 29,049-image multi-source corpus centered on the reinforced-concrete urban stock of Kazakhstan, including a 12-month, fixed-viewpoint sequence of 2255 frames that isolates invariance to construction stage, illumination, weather, and season. We formalize a reproducible annotation protocol for three recurring structural ambiguities—incomplete upper floors, rooftop superstructures, and open ground-level pilotis—and propose DINOv2-MSTS, a dual-branch architecture that aggregates multi-scale patch-token statistics from a frozen self-supervised backbone, trained with an Ordinal-Aware Annotation-Uncertainty (OAU) loss for which its Gaussian spread is learned rather than fixed. On the 5359-image Korter + Mendeley 21-category benchmark, the model attains 80% top-1 accuracy, 94% within $\pm 1$ floor accuracy, and 0.28-floor mean absolute error on this saturated 21-category task (a lower bound for buildings of 21 or more floors) using only 1.84 M trainable parameters, 165× fewer than a fully fine-tuned Vision Transformer, which it outperforms by eight accuracy points. On the separate 2255-frame IITU fixed-label robustness probe, it preserves the correct six-floor prediction in 91% of frames (0.09-floor MAE). The corpus, protocol, architecture, and loss together provide a reproducible benchmark for construction-stage building monitoring. Full article

This study examines the context-dependent association between eye-level motor traffic and metro-integrated cycling in Shenzhen, China. To address the limitations of static street-view imagery, we constructed a traffic-informed AIGC-enhanced analytical framework to approximate peak-hour visual motor-traffic conditions. The resulting eye-level motor-traffic measure was incorporated into OLS, GWR, and MGWR models together with land-use, road-network, development-intensity, and streetscape variables. The results show that this measure was positively associated with metro-integrated cycling volume primarily during the weekday morning peak, while the association weakened or became statistically insignificant during evening and weekend periods. We describe this pattern as a commuter’s paradox-like association: visible motor traffic may co-occur with high first-/last-mile cycling demand in high-intensity commuting environments, rather than necessarily deterring cycling. The analysis further suggests a threshold-like land-use pattern in which residential density may act as a background precondition rather than a linear driver during peak hours. This study illustrates the methodological applicability of Geospatial Artificial Intelligence (GeoAI) for addressing static-data limitations and provides planning implications for evaluating station-area feeder cycling environments. Full article

Contemporary buildings and urban spaces are increasingly expected to support psychological well-being—a quality often termed “restorativeness.” Conventional approaches to quantifying restorativeness rely on subjective surveys or coarse green metrics, failing to capture how specific building morphologies and street-level visual configurations shape restorative experiences, particularly for stress-prone groups such as young adults. This study develops a deep-learning-driven framework linking building visual elements to youth-specific perceived restorativeness, using Changsha, China, as a testbed. The framework comprises three AI-powered modules: the TrueSkill algorithm trains a deep learning model to predict six dimensions of youth perception (e.g., beautiful, clean, safe) from pairwise comparisons of street view images; the Mask2Former architecture segments street-level imagery into 18 building and street attributes; and the XGBoost-SHAP pipeline uncovers nonlinear associations and threshold-like patterns between these attributes and the composite Built Environment Restorativeness Index (BERI). Results reveal three key insights: tree coverage shows a sustained positive association without saturation; building density exhibits a weakening association at high levels, suggesting possible saturation; and road proportion follows a bidirectional pattern, shifting from negative to positive beyond a certain range. Spatially, high BERI zones concentrate where ecological assets and diverse building functions co-occur, while youth perception exhibits systematic mismatches (e.g., “beautiful but not clean,” “safe but not lively”), traceable to imbalances in building form, street furniture, and commercial mix. These findings advance AI-assisted evaluation of built environments by shifting from one-dimensional metrics to interpretable, design-relevant diagnostics, offering a replicable evidence base for crafting youth-responsive buildings and streets. Full article

This study investigates the association between urban greenness and sales in commercial alleys. We focus on 1090 commercial alleys in Seoul, South Korea, defined as neighborhood-scale open commercial streets or districts composed of small retail, service, cafe, and restaurant businesses, and combine spatially explicit measures of greenness with data on weekend sales to assess how variation in vegetation is associated with local economic performance. Greenness is measured by the normalized difference vegetation index (NDVI) derived from remote sensing imagery. We employ a set of global and spatially explicit models, including Ordinary Least Squares (OLS), Geographically Weighted Regression (GWR), Multiscale Geographically Weighted Regression (MGWR), and a Python Geographical Random Forest (PyGRF, v0.0.12), to capture both overall and location-specific relationships. The results show that higher levels of greenness are significantly associated with higher weekend sales, with spatial heterogeneity observed across different areas of the city. The green investment efficiency index (GIEI) results further identify clusters of high investment efficiency in areas characterized by strong greenness–sales associations and relatively limited existing greenness. High GIEI values were concentrated in areas near natural amenities and dense residential neighborhoods, indicating potential priority locations for targeted greening interventions. By linking objective measures of greenness to observed sales at the scale of everyday commercial environments, this study contributes to a better understanding of how urban greenness is associated with consumer behavior and local economic activity. The findings provide practical implications for identifying areas where greening strategies may be considered as part of broader efforts to support more resilient and sustainable neighborhood commercial areas, while recognizing that the observed relationships are associative rather than causal. Full article

Extracting roads accurately from remote sensing images is important for map updates, traffic analysis, and infrastructure monitoring. Medium-resolution multispectral images can provide useful surface and background information, but when used alone, the spatial details are limited for retaining narrow roads, intersection structures, and fine road topologies. To address this problem, this paper proposes GeoRoad-UPerNet, a Geo-1-centered weakly supervised multispectral framework for road extraction. In this framework, Geo-1 serves as the primary 16-band multispectral source, Sentinel-2 Level-2A imagery serves as auxiliary contextual support, and OpenStreetMap (OSM) road information is converted into proxy supervision rather than dense manual ground truth. GeoRoad-UPerNet contains three modules: a Geo Spectral Semantic Stem (GSSS), a Geo-Auxiliary Gated Fusion module (GAGF), and a Road Semantic Multi-Task Head (RSMH). GSSS strengthens road-sensitive multispectral responses in the Geo-1 branch. GAGF injects Sentinel-2 context through a Geo-centered gate instead of symmetric channel concatenation. RSMH imposes restrained hierarchy- and material-aware semantic regularization on the shared decoder representation during training. On the fixed source-domain benchmark, the complete model achieves an IoU of 0.7204, an F1-score of 0.8375, a Precision of 0.8092, and a Recall of 0.8678 against OSM-derived proxy masks. Relative to the UPerNet-MiT-B3 early-fusion baseline, IoU, F1-score, and Precision increase by 6.29%, 3.65%, and 12.58%, respectively. These results indicate that role-aware multisource organization improves road extraction under proxy supervision and reduces boundary noise and background false positives. Full article

Semantic segmentation of street-view imagery has become an important computational tool for urban morphological analysis and the evaluation of street spatial quality. However, existing methods still struggle in complex urban environments. Major challenges include large variations in building façade scales, degradation of boundary information, and severe class imbalance. These issues limit the ability of current models to capture structurally meaningful urban forms. To address these challenges, this study proposes a high-resolution street-view segmentation framework, termed HieraWaveSeg. The model aims not only to improve pixel-level segmentation accuracy but also to enhance the interpretability of urban morphology through structured representations of street space. Specifically, a Hiera Transformer backbone is employed to capture hierarchical spatial semantics. A Path Aggregation Network is further introduced to strengthen cross-scale feature interaction and improve structural consistency in complex scenes. In addition, a Wave Fusion module based on the Haar wavelet transform is incorporated to preserve fine-grained architectural details by enhancing high-frequency boundary and texture information during decoding. Unlike conventional segmentation approaches that primarily focus on object recognition, this study introduces a morphology-oriented semantic reconfiguration strategy. This strategy reorganizes original categories into functionally meaningful urban units. As a result, the segmentation outputs can be more directly linked to urban morphological indicators, such as façade continuity, spatial enclosure, and interface permeability, thereby improving interpretability in architectural and urban design contexts. To further address class imbalance, a composite loss function combining weighted cross-entropy and Dice loss is adopted, together with a median frequency balancing strategy. Experimental results on the CamVid and Cityscapes datasets demonstrate that the proposed method consistently outperforms several state-of-the-art baselines in both segmentation accuracy and structural preservation. Beyond quantitative improvements, the results indicate that the proposed framework generates more coherent and morphologically meaningful urban representations, supporting further quantitative analysis in urban morphology and architectural studies. Full article