Analyzing the typology and livability of 15-minute travel at metro stations in high-density cities: A case study of Singapore

Cities, Journal Year: 2025, Volume and Issue: 158, P. 105727 - 105727

Published: Jan. 17, 2025

Language: Английский

---

Comprehending the interaction between urban function and morphology at traffic analysis zones scale: The case study from Hangzhou

Geographical Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Abstract Urbanisation is transitioning from disorderly sprawl to compact intensification, accompanied by functional differentiation and morphological changes spatially. This study addresses the relationship between urban functions morphologies at block scale in Hangzhou. Leveraging geo‐big data, we adopt a points of interest (POI) weighting method map four essential functions—residential, commercial, public service, industrial—at traffic analysis zones (TAZ) scale. Additionally, estimate indices using building footprint data volume data. Our investigation reveals intriguing patterns: residential, service exhibit central concentration trend diminishing towards periphery, whereas industrial demonstrate multi‐hotspot distribution. Morphological like patch density mean while size shape index, presenting pronounced peripheral distribution trend. Significantly, nuanced associations were elucidated. Residential tend display dense small patches, commercial areas showcase larger volumes, complex shapes. Furthermore, construction intensity‐based heterogeneity unveils dynamics morphologies, particularly high‐density areas. These findings underscore importance integrating considerations into planning, offering fresh perspective for zoning planning.

Language: Английский

---

Visual Feature Extraction and Tracking Method Based on Corner Flow Detection

IECE transactions on intelligent systematics., Journal Year: 2024, Volume and Issue: 1(1), P. 3 - 9

Published: May 15, 2024

Frontend feature tracking based on vision is the process in which a robot captures images of its surrounding environment using camera while motion. Each frame image then analyzed to extract points, are subsequently matched between pairwise frames estimate robot’s pose changes by solving for variations these points. While matching methods that rely descriptor-based approaches perform well cases significant lighting and texture variations, addition descriptors increases computational costs introduces instability. Therefore, this paper, novel approach proposed combines sparse optical flow with Shi-Tomasi corner detection, replacing use descriptors. This new method offers improved stability situations challenging maintaining lower costs. Experimental results, validated OpenCV library Ubuntu operating system, provide evidence algorithm’s effectiveness efficiency.

Language: Английский

---

Interpreting core forms of urban morphology linked to urban functions with explainable graph neural network

Computers Environment and Urban Systems, Journal Year: 2025, Volume and Issue: 118, P. 102267 - 102267

Published: Feb. 24, 2025

Language: Английский

---

Formalising the urban pattern language: A morphological paradigm towards understanding the multi-scalar spatial structure of cities

Cities, Journal Year: 2025, Volume and Issue: 161, P. 105854 - 105854

Published: March 5, 2025

Language: Английский

---

Assessment of the impact of urban block morphological factors on carbon emissions introducing the different context of local climate zones

Sustainable Cities and Society, Journal Year: 2024, Volume and Issue: unknown, P. 106073 - 106073

Published: Dec. 1, 2024

Language: Английский

---

Exploring road safety using alignment perspective features in real driving images: A case study on mountain freeways

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(6), P. e0305241 - e0305241

Published: June 17, 2024

Introduction While driving, drivers frequently adapt their driving behaviors according to perception of the road’s alignment features. However, traditional two-dimensional methods lack ability capture these features from driver’s perspective. Method This study introduces a novel method for road recognition, employing image recognition technology extract perspective features, namely skewness (APS) and kurtosis (APK), in-real images. Subsequently, K-means clustering algorithm is utilized segment classification based on APS APK indicators. Various sliding step length are employed, with ranging 100m 400m. Furthermore, accident rates different clusters analyzed explore relationship between traffic safety. A 150 km mountain section Erlianhaote-Guangzhou freewway Huaiji Sihui selected as case study. Results The results demonstrate that using criteria produces favorable outcomes, superior performance achieved shorter lengths fewer cluster centers. reveals notable differences in across categories; while parameters-based fail differences. most significant categories observed 100m, significance gradually diminishing increases disappearing entirely when exceeds 300m. Implication These findings validate reliability (APS APK) safety analysis, providing valuable insights management.

Language: Английский

---

Towards a Deep Automatic Generation of Figure-ground Maps

ISPRS annals of the photogrammetry, remote sensing and spatial information sciences, Journal Year: 2024, Volume and Issue: X-4/W5-2024, P. 33 - 39

Published: June 27, 2024

Abstract. Figure-ground maps play a key role in many disciplines where urban planning or analysis is involved. In this context, the automatic generation of such with respect to certain requirements and constraints an important task. This paper presents first step towards deep figure-ground built density generated scenes controlled taken into account. preformed building upon Geographic Data Translation model which has been applied generate less available geospatial features, e.g. footprints, from more widely data, street network using conditional Generative Adversarial Networks. A novel processing approach introduced incorporate population accordingly. Furthermore, impact both level detail network, i.e. its sparsity density, spatial resolution training data on investigated. The qualitative results reveal obvious these parameters layout unbuilt areas. Our paves way for expansion existing districts by future neighbourhoods considering factors as further will be subject work.

Language: Английский

---

Quantifying spatial patterns of urban building morphology in the China’s Guangdong-Hong Kong-Marco greater bay area

International Journal of Digital Earth, Journal Year: 2024, Volume and Issue: 17(1)

Published: Aug. 19, 2024

Understanding the spatial patterns of urban building morphology is crucial for revealing interplay between built and social environments. Previous research has predominantly concentrated on computation building-level metrics which makes it challenging to quantify compare variations across different cities. Using newly available world settlement footprint 3D (WSF3D) data, this study examines various cities within Guangdong-Hong Kong-Macao Greater Bay Area, a rapidly urbanizing region in China. Specifically, we applied concentric ring approach delineate gradients fraction, area, height, volume from center suburban fringes. Subsequently, utilized dynamic time warping multi-dimensional scaling technique facilitating comparative analysis these Developed demonstrated more homogenous distributions morphologies; however, notable differences were observed among distinct metrics. Furthermore, correlation degree development was revealed, suggesting that developed exhibit significantly smaller declines core rural periphery.

Language: Английский

---

Extracting Meso- and Microscale Patterns of Urban Morphology Evolution: Evidence from Binhai New Area of Tianjin, China

Land, Journal Year: 2024, Volume and Issue: 13(11), P. 1735 - 1735

Published: Oct. 23, 2024

Understanding and recognizing urban morphology evolution is a crucial issue in planning, with extensive research dedicated to detecting the extent of expansion. However, as development patterns shift from incremental expansion stock optimization, related studies on meso- microscale face limitations such insufficient spatiotemporal data granularity, poor generalizability, inability extract internal patterns. This study employs deep learning meso-/microscopic form indicators develop generic framework for extracting describing meso-/microscale morphology. The includes three steps: constructing specific datasets, semantic segmentation form, mapping using Tile-based Urban Change (TUC) classification system. We applied this conduct combined quantitative qualitative analysis Binhai New Area 2009 2022, detailed visualizations at each time point. identified that different locations area exhibited seven distinct patterns: edge areal expansion, preservation developmental potential, industrial land pattern, rapid comprehensive demolition construction linear mixed evolution, stable evolution. results indicate phase, high-density areas exhibit multidimensional characteristics by region, period, function. Our work demonstrates potential grid providing scalable, cost-effective, quantitative, portable approach historical understanding.

Language: Английский

---