Embedded Files
1. Estimating Tree, Building, and Sky View Factor using Google Street View
1. Estimating Tree, Building, and Sky View Factor using Google Street View
February 2018
Workflow of semantic scene parsing using PSPNet.
Workflow for View factor calculations using GSV images proposed by Gong et al. (2018).
Maps of GSV-based SVF in (a), TVF in (b), and BVF in (c) of street canyons in high-density urban areas of Hong Kong derived from 29,264 GSV images along streets at 30m intervals; (d) Frequency density of SVF (blue), TVF (green), and BVF (grey).
This study aims at
(1) Developing an approach for automatically and accurately deriving sky, tree, and building view factors of street canyons in the high-density urban environment of Hong Kong using the publicly available Google Street View (GSV) images and a deep-learning feature extraction algorithm for extraction of street features (sky, tree, and building);
(2) Verifying the accuracy of the developed GSV-based method using reference data of hemispheric photography from field survey;
(3) Comparing the GSV-based with 3D-GIS-based view factors estimates and investigating the impact factors for the discrepancies between them.
As a result, maps of SVF, TVF, and BVF of the street canyon in high-density urban areas of Hong Kong are generated. Validation using reference data of hemispheric photography from field surveys in compact high-rise and low-rise areas shows that the GSV-based view factor estimates have satisfying agreements (all with R2 >0.95) with the reference data, suggesting the effectiveness and high accuracy of the developed method. This is the first reported use of hemispheric photography for direct verification in a GSV-based streetscape study. Furthermore, a comparison between GSV-based and 3D-GIS-based SVFs shows that
(1) The two SVF estimates are significantly correlated (R2=0.40, p<0.01), and show better agreement in high-density areas. However, the latter overestimates SVF by 0.11 on average;
(2) The differences between them are significantly correlated with street trees (R2=0.53): the more street trees, the larger the difference. This suggests that a misrepresentation of street trees in a 3D-GIS model of a street environment is the dominant factor contributing to the large discrepancies between the two datasets.
The developed GSV-based method for analyzing view factors in a 3D street environment makes the large-scale sky, tree, building view factors estimation possible.
Gong, F.-Y.*, Zeng, Z.-C., Zhang, F., Li, X., Ng, E., & Norford, L. K. (2018). Mapping sky, tree, and building view factors of street canyons in a high-density urban environment. Building and Environment, 134, 155-167.
2. Estimating Global, Direct, and Diffuse Radiation using Google Street View
2. Estimating Global, Direct, and Diffuse Radiation using Google Street View
October 2018
Physical basis of incident solar radiation at street level.
Workflow for solar radiation calculation using GSV images.
Maps of monthly means of daily solar radiaiton (MJ/m2/day) in street canyons are averaged over 6 years from 2009 to 2014.
This study aims at
(1) Developing an approach for accurately calculating the street-level global, direct, and diffuse solar irradiance in high-density urban environments using publicly available GSV images;
(2) Quantifying the spatial and temporal patterns of street-level solar irradiation in the high-density urban areas of Hong Kong;
(3) Investigating the impacts of street canyon geometries (street orientation and aspect ratio) and morphologies (sky opening and obstructions by buildings and trees) on street-level solar irradiation.
As a result, maps of global, direct, and diffuse radiations of the street canyon in high-density urban areas of Hong Kong are generated. Verifications of our developed method using free-horizon observatory from HKO and field measurements in a high-density street canyon show that both the clear-sky (without cloud effects) and all-sky (with cloud effects) solar irradiance of street canyons accurately capture the diurnal and seasonal cycle in high-density environments.
The developed method in this study to map the street-level solar radiation can be, on a global scale, applied to cities with available coverage of GSV images. This method provides a low-cost and effective streetscape mapping approach for urban studies.
Gong, F.-Y.*, Zeng, Z.-C., Ng, E., & Norford, L. K. (2018). Spatiotemporal pattern of street-level solar irradiance estimated using Google Street View in a high-density urban environment. Building and Environment, 148, 547-566.
3. Estimating Spatial Accessibility of Green Infrastructure
3. Estimating Spatial Accessibility of Green Infrastructure
September 2016
Accessibility analysis of urban green space patches in terms of walking distance, by increasing the radius of buffer region around green space patches from 0m-5000m, including 6 categories (0-100m, 100-300m, 300-500m, 500-1000m, 1000-3000m and 3000-5000m).
Accessibility analysis of urban green space patches in terms of walking time, by assuming an average walking speed of elderly, 0.88m/s, including 6 categories (0-2min, 2-5min, 6-10min, 10-20min, 20-60min and 60-100min).
Hong Kong is an ageing society. According to the World Health Organization, an estimated 22% of its residents will be 65 years old or above by the year 2030. Physical activities provide an important way for older people to stay healthy. Green space is recognized as an important environmental setting for physical activity. Therefore, improving elderly accessibility to green space is useful in promoting more physical activity among them.
The aim of this study is to modelling and assessing the elderly accessibility to urban green spaces in Hong Kong, by
(1) modelling urban green spaces available for elderly people using the landscape fragmentation index (LFI) of green space patches;
(2) accessibility analysis in terms of both distance and time for elderly people to get urban green space patches.
As a result, regions with LFI from high to low and accessibility from good to poor are Kowloon, Hong Kong Island, and New Territories. Kowloon has the highest LFI and accessibility. However, it is limited by small size green space patches which may not be in good quality and attractive for elder people. New Territories, due to its large number of available lands, has the largest proportion of big size green space patches. However, they are limited by their poor accessibility. Based on this evidence, we would suggest the prior improvement (1) in Kowloon is to have more green spaces with larger sizes, (2) in New Territories is to increase the green space accessibility either by improving the walking routes or building more small to medium size green space patches that can well cover the whole region.
Gong, F.-Y.*, Zeng, Z.-C., & Ng, E. (2016). Modeling elderly accessibility to urban green space in High-density cities: A case study of Hong Kong. Procedia Environmental Sciences, 36, 90-97.
4. Urban Climate Map in High-density Cities of Macau
4. Urban Climate Map in High-density Cities of Macau
August 2015
GIS-based Existing Analysis of Wind Environment
Wind Environment Evaluation of Macau Existing Condition using Prognostic-Diagnostic Hybrid mode (WRF-CALMET)
Micro-scale CFD Simulation of New Reclamation Areas
This study developed a web-based platform for interactive assessment of 10 years hourly wind information over the Macau area from 2000 to 2010. The Prognostic-Diagnostic Hybrid mode (WRF-CALMET) wind field downscaling is used for generating the very high-resolution wind map (down to 100m resolution).
This study investigates the pedestrian level ventilation in Macau using the parallelized large-eddy simulation model PALM.
The aims of the study are:
(1) to analyze the ventilation in Macau’s present city districts and in the planned reclamation areas,
(2) to evaluate the impact of the planned reclamation areas on the ventilation within the present city districts,
(3) to investigate how urban morphological quantities.
The evaluation result of wind environment show that:
(1) Planned reclamation areas A and B have the potential to affect the ventilation in the present city districts of the Peninsula de Macau;
(2) In case of south-east wind, zone A and zone B cause decreased ventilation in few districts downstream;
(3) In case of south wind, zone B has a strong impact on the ventilation of the peninsula, with better ventilation downstream of the western part of zone B and worse ventilation downstream of the eastern part of zone B;
(4) In case of north wind, zones C, D and E1 significantly reduce the ventilation in many districts of Taipa;
(5) Most of the planned reclamation areas themselves will be well-ventilated, regardless of the wind direction. Only in zone A will be some parts with bad to moderate ventilation due to the high value of ground coverage ratio and frontal area density.
5. Study of the Methods of Urban Green Space Planning in High-density Cities
5. Study of the Methods of Urban Green Space Planning in High-density Cities
October 2011
Based on the features of high-density urban green space in Macau, this study first set up green space function-oriented ‘Standard for Classification of Urban Green Space’. Then the analysis of two green indexes: Fragmentation of Green Space Patch and Recreation-Residential Land Rate (RLR), calculated the density of green space layout and recreation-residential land space layout respectively.
After status analysis, the study employed travel cost methodology of GIS and built the cost distance mode of green space accessibility, and finally generated ‘Accessibility Grade Layout Map of Recreation Green Space in Macau’. It can support the scientific reference to the poor grade service blind area of green space and improve service capabilities of recreational green space. Based on analysis of human recreation behavior, the study generated the high-density green space layout planning method to improve the usage efficiency and fairness of green space for human.
Google Sites
Report abuse