Advances in Building Energy Research, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23
Published: Feb. 3, 2025
Language: Английский
Advances in Building Energy Research, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23
Published: Feb. 3, 2025
Language: Английский
Journal of Cleaner Production, Journal Year: 2025, Volume and Issue: unknown, P. 144768 - 144768
Published: Jan. 1, 2025
Language: Английский
Citations
1Remote Sensing, Journal Year: 2023, Volume and Issue: 15(15), P. 3840 - 3840
Published: Aug. 1, 2023
Identifying the main factors influencing land surface temperature (LST) of each local climate zone (LCZ) built type is great significance for controlling LST. This study investigated LST LCZ in two Asian megacities: Tokyo and Shanghai. Each area both megacities was classified according to scheme. The diurnal LST, pervious fraction (PSF), albedo (SA), average building height (⟨BH⟩), gross coverage ratio (λp) were also calculated. Finally, influence properties on investigated. results demonstrated that different types differed ⟨BH⟩ factor compact mid-rise open high-rise Tokyo, Shanghai; PSF other types. Moreover, negatively correlated with Based above characteristics type, specific mitigation strategies proposed approach this can contribute perspectives urban planners policymakers develop highly feasible reasonable strategies.
Language: Английский
Citations
17International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2024, Volume and Issue: 127, P. 103696 - 103696
Published: Feb. 2, 2024
Urban scene understanding and functional identification are essential for accurately characterizing the spatial structure optimizing city layouts during rapid urbanization. Multimodal data is important recognizing distribution patterns of urban functions revealing internal details. Previous studies have focused primarily on remote sensing imagery points interest (POIs) data, overlooking role building characteristics in determining scenes. These also limited terms mining fusing multimodal features. To address these challenges, this study proposes a fusion framework that integrates imagery, POIs, footprints mapping. The employs dual-branch model extracts visual semantic features from socioeconomic auxiliary such as POIs footprints. A branch attention module designed to assign weights Additionally, multiscale feature introduced extract combine through modal interaction. Experiments Beijing Chengdu validate effectiveness proposed with overall accuracy 90.04% 92.07%, kappa coefficient 0.881 0.895, respectively. This provides empirical evidence support accurate planning further promote sustainable development. source code at: https://github.com/sssuchen/MMFF.
Language: Английский
Citations
8IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Journal Year: 2024, Volume and Issue: 17, P. 9728 - 9744
Published: Jan. 1, 2024
Clarifying the factors that influence land surface temperature (LST) is crucial for proposing specific LST mitigation strategies. This study focuses on Beijing-Tianjin-Hebei (BTH) Region and investigates influencing of various local climate zone (LCZ) built types from perspectives urban morphology, cover, human activity. The results suggest areas LCZ vary across cities within BTH Region, attributed to differences in city size Gross Domestic Product (GDP). area Beijing Tianjin, with significantly high sizes GDP, exceeds 2000 km2. In contrast, Qinhuangdao, Zhangjiakou Chengde, which have relatively low this less than 500 However, main same type are highly consistent. Building coverage ratio (BCR), average building height (ABH) pervious fraction (PSF) three most important factors. correlation between BCR mainly concentrated compact high-rise open types, Pearson coefficient (r) ranging 0.2 0.44; ABH high-rise, mid-rise, mid-rise r -0.2 -0.52; PSF almost all -0.56. By integrating these findings features each strategies were further proposed. can help develop context Coordinated Development thereby promoting healthy sustainable development region.
Language: Английский
Citations
8Remote Sensing of Environment, Journal Year: 2024, Volume and Issue: 306, P. 114119 - 114119
Published: March 21, 2024
Language: Английский
Citations
7Cities, Journal Year: 2024, Volume and Issue: 150, P. 104999 - 104999
Published: April 14, 2024
Language: Английский
Citations
7IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2024, Volume and Issue: 62, P. 1 - 14
Published: Jan. 1, 2024
Local climate zone (LCZ) mapping can explore the variability of impact urban form on thermal environment in different contexts, and large-scale LCZ help us to better understand spatial temporal dynamics areas around world. Studies have indicated that deep learning-based methods effectively perform classification. However, accuracy classification datasets is still unsatisfactory, mainly due fact traditional convolutional neural networks are not good at mining contextual information, which crucial for fully understanding remote sensing scenes. In this paper, solve problem, we propose an method based images by coupling multi-level features mined from global local ranges with prior knowledge, named LCZ-MFKNet. The extracted through Swin Transformer space-maintained ResNet (SM-ResNet) model branches, respectively, then fused improved squeeze-and-excitation (iSE) module. knowledge studied theoretical definition experimental tests two typical sets categories easily confounded multi-class but separable two-class Experiments conducted large publicly available So2Sat LCZ42 dataset, where proposed LCZ-MFKNet achieved highest accuracy. Moreover, six megacities were selected globally mapping, results verified general applicability mapping.
Language: Английский
Citations
6Scientific Data, Journal Year: 2024, Volume and Issue: 11(1)
Published: Feb. 13, 2024
Abstract Urbanization has altered land surface properties driving changes in micro-climates. Urban form influences people’s activities, environmental exposures, and health. Developing detailed unified longitudinal measures of urban is essential to quantify these relationships. Local Climate Zones [LCZ] are a culturally-neutral classification scheme. To date, LCZ maps at large scales (i.e., national, continental, or global) not available. We developed an approach map LCZs for the continental US from 1986 2020 100 m spatial resolution. lightweight contextual random forest models using hybrid model development pipeline that leveraged crowdsourced expert labeling cloud-enabled modeling – could be generalized other countries continents. Our achieved good performance: 0.76 overall accuracy (0.55–0.96 class-wise F1 scores). our knowledge, this first high-resolution, US. work may useful variety fields including earth system science, planning, public
Language: Английский
Citations
6International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2024, Volume and Issue: 131, P. 103944 - 103944
Published: June 3, 2024
In the last decade, several methods have been developed for Local Climate Zone (LCZ) mapping, encompassing Remote Sensing and Geographic Information Systems (GIS) −based procedures. Combined approaches also proposed to compensate intrinsic limitations that characterized their separate application. Recent work has disclosed potential of hyperspectral satellite imagery improving LCZ identification. However, use data mapping is yet be fully unfolded. A combined GIS-based method exploit integration PRISMA multispectral Sentinel-2 images with ancillary urban canopy parameter layers. Random Forest algorithm applied feature sets obtain classification. The tested on Metropolitan City Milan (Italy), period from February August 2023. spectral separability analysis carried out investigate improvement in identification using comparison data, as well improvements pan-sharpened images. resulting maps' quality evaluated by extracting accuracy metrics performing inter-comparisons maps computed Generator benchmark tool. Inter-comparisons yield promising results a mean Overall Accuracy increase 16% each class. Furthermore, we find improves detection LCZs compared Sentinel-2, 5%, line higher signatures training samples.
Language: Английский
Citations
6Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 99, P. 104863 - 104863
Published: Aug. 15, 2023
Language: Английский
Citations
16