From Heat Resilience to Sustainable Co-Benefits: Adaptive Urban Morphology Generation based on Multimodal Data Fusion and a Novel Generative Framework

Sustainable Cities and Society, Год журнала: 2025, Номер unknown, С. 106452 - 106452

Опубликована: Май 1, 2025

Язык: Английский

---

Mapping Street Patterns with Network Science and Supervised Machine Learning

ISPRS International Journal of Geo-Information, Год журнала: 2024, Номер 13(4), С. 114 - 114

Опубликована: Март 28, 2024

This study introduces a machine learning-based framework for mapping street patterns in urban morphology, offering an objective, scalable approach that transcends traditional methodologies. Focusing on six diverse cities, the research employed supervised learning to classify networks into gridiron, organic, hybrid, and cul-de-sac with street-based local area (SLA) as unit of analysis. Utilising quantitative metrics GIS, analysed form through random forest method, which reveals predictive features enables deeper understanding spatial structures cities. The findings showed distinctive structures, such ring formations cores, indicating stages development socioeconomic narratives. It also analysis has major impact identification patterns. Concluding is critical tool suggests future studies should expand this include more cities elements. would enhance modelling growth inform sustainable, human-centric planning. implications are significant policymakers planners seeking harness data-driven insights

Язык: Английский

---

Generalized spatio-temporal-spectral integrated fusion for soil moisture downscaling

ISPRS Journal of Photogrammetry and Remote Sensing, Год журнала: 2024, Номер 218, С. 70 - 86

Опубликована: Окт. 19, 2024

Язык: Английский

---

Synthesis and Detection Algorithms for Oblique Stripe Noise of Space-Borne Remote Sensing Images

IEEE Transactions on Geoscience and Remote Sensing, Год журнала: 2024, Номер 62, С. 1 - 14

Опубликована: Янв. 1, 2024

Oblique stripe noise widely appears in remote sensing images after image correction, exhibiting arbitrary tilt angles and parallel distribution. Due to its randomness lengths, oblique increases the difficulty of detection compared vertical or horizontal noise. For first time, we propose a group synthesis algorithms combining imaging mechanisms deep learning. To get controllable synthetic data for training model, two sample augmentation methods are presented by correction's with new linear transformation generative adversarial network algorithm Cycle-GAN, respectively. A large-scale simulated dataset (SOSD, dataset) is using these methods. learning (RDOS, Robust Noise) considering presence RDOS trained both SOSD real dataset, it obtains optimal model testing. The experimental results show that accuracy reaches 82.93%, recall rate 85.17%, F1 score 84.04%, average precision (AP) 82.34%, frames per second (FPS) 33.33. Compared general line models, our exceeds ~300% ~60% speed. In future, proposed have great potential application various areas such as quality evaluation, preprocessing, engineering problems related multi-angle object detection.

Язык: Английский

---

From the abundance perspective: Multi-modal scene fusion-based hyperspectral image synthesis

Information Fusion, Год журнала: 2024, Номер 108, С. 102419 - 102419

Опубликована: Апрель 10, 2024

Язык: Английский

---

Pano2Geo: An efficient and robust building height estimation model using street-view panoramas

ISPRS Journal of Photogrammetry and Remote Sensing, Год журнала: 2024, Номер 215, С. 177 - 191

Опубликована: Июль 11, 2024

Building height serves as a crucial parameter in characterizing urban vertical structure, which has profound impact on sustainable development. The emergence of street-view data offers the opportunity to observe 3D scenarios from human perspective, benefiting estimation building height. In this paper, we propose an efficient and robust model, call Pano2Geo by precisely projecting panorama (SVP) coordinates geospatial coordinates. Firstly, SVP refinement stratagem is designed, incorporating NENO rules for observation quality assessment four aspects: number buildings, extent nodes, orthogonal observations, followed application art gallery theorem further refine SVPs. Secondly, model constructed, provides pixel-level projection transformation locating features buildings SVP. Finally, valid feature points are extracted based slope mutation test, projected using so obtain proposed was evaluated city Wuhan China, results indicate that can accurately estimate height, with average error 1.85 m. Furthermore, compared three state-of-the-art methods, shows superior performance, only 10.2 % have absolute errors exceeding 2 m, Map-image-based (27.2 %), Corner-based (16.8 Single-view-based (13.9 %) methods. method achieves optimal less than 50 existing SVPs, leading highly estimation, particularly areas high density. Moreover, exhibits robustness maintaining within m even shape complexity occlusion degree increase Our source dataset code available at https://github.com/Giser317/Pano2Geo.git.

Язык: Английский

---

UnmixDiff: Unmixing-based Diffusion Model for Hyperspectral Image Synthesis

IEEE Transactions on Geoscience and Remote Sensing, Год журнала: 2024, Номер 62, С. 1 - 18

Опубликована: Янв. 1, 2024

Язык: Английский

---

A Study on Urban Block Design Strategies for Improving Pedestrian-Level Wind Conditions: Cfd-Based Optimization and Generative Adversarial Networks

Опубликована: Янв. 1, 2023

Urban block layout design presents a critical and challenging task for urban planners, as directly influences the physical structure of area, interaction between people their environment, microclimate in areas. Thus, developing appropriate strategies plays decisive role this process. This research uses generative adversarial networks (GAN) technique to explore strategies. The GAN can effectively generate real diverse blocks based on learning existing morphological properties. Therefore, it be powerful approach research. Genetic algorithms not only help identify ideal solution but also enable summary key by numerous evolutionary periods. study CFD-based optimization framework that utilizes genetic algorithm 3D models generated improve wind conditions at scale. By comparing solutions, successfully optimized three objectives 68.36%,51.74%, 41.83%. Ridge regression examined relationship objective functions indices. maximum R2 value ridge reached 0.801, indicating predict indicators. Design wind-friendly were developed validated case studies. suggest architects should prioritize building layout, height boundary areas, shape, these factors significantly affect outdoor conditions.

Язык: Английский

---

Map Diffusion - Text Promptable Map Generation Diffusion Model

Опубликована: Ноя. 13, 2023

This paper introduces a novel text promptable map generation model, leveraging recent advancements in generative models. Promptable has broad applications, democratizing access to geographic data, enhancing decision-making, improving communication, and enabling customization. Map Diffusion generates maps based on textual descriptions, allowing users describe region, the model corresponding map. We conduct comprehensive review of related work, highlighting unique contributions our model. also provide insights into dataset creation, architecture, training procedures, experimental results. research marks significant step harnessing models for generation, opening doors future exploration this field.

Язык: Английский

---

Georeferencing for Planning Underground Networks

Опубликована: Янв. 27, 2024

Often the problems related to planning of electrical networks, drinking water (sewage), data, and transportation, use ideal models that are not geo-referenced.However, present work exposes a suitable innovative methodology achieve data analytics from OSM files.This geographic information file can be freely downloaded https : //www.openstreetmap.org/[1].Then, applying geo-referenced model network provides characteristics environment in x y coordinates.In this sense, using longitude latitude allows location with less error for distance calculations used optimal wireless sensors (smart meters), transformers, determination route burying power lines, as well determine population growth specific area, evaluation rescue zone or geographical areas vaccination campaigns, among other options, [2]- [4].Over time, scientific publications seek innovate concerning previous by community.Thus, applications distribution networks based on perform modeling deployment underground networks.Simulation tools such Cymdist [5]- [7] generally evaluate deployed model.In way, aim is serve minimum cost electric cable considering variables voltage drops [8], [9].

Язык: Английский

---