| id |
sigradi2023_387 |
| authors |
Dong, Jiahua, Lin, Shuiyang and van Ameijde, Jeroen |
| year |
2023 |
| title |
Predicting Network Integration Based on Satellite Imagery Around High-Density Public Housing Estates Through Machine Learning |
| source |
García Amen, F, Goni Fitipaldo, A L and Armagno Gentile, Á (eds.), Accelerated Landscapes - Proceedings of the XXVII International Conference of the Ibero-American Society of Digital Graphics (SIGraDi 2023), Punta del Este, Maldonado, Uruguay, 29 November - 1 December 2023, pp. 795–806 |
| summary |
In studies focusing on environmental and health aspects of urban planning, the integration of road networks within the built environment emerges as an important metric for assessing the livability and healthiness of neighborhoods. The complexity and diversity of the road networks are significant for shaping vibrant streets. In Hong Kong’s ongoing construction program of large-scale public housing estates, the design prioritizes the connectivity of pedestrian circulation to foster social interaction among residents and encourage the utilization of recreational facilities. In this study, an analytical framework is developed to interpret public housing estate spatial layout based on satellite imagery. It extracts road networks using neural networks and vectorizes results to analyze network integration around estates to predict social interactions. The aim of this process is to employ a machine learning workflow to analyze options for newly planned estates, where the design configuration can be further optimized based on its potential to stimulate social engagement and community interaction. Due to the scalability and universality of the method, the research can contribute to improved road networks and sociable housing complexes in Hong Kong, or in other international cities of similar density and vibrancy. |
| keywords |
Network Integration, Spatial Structure, Satellite Imagery, Machine Learning, Hong Kong Public Housing |
| series |
SIGraDi |
| email |
|
| full text |
 file.pdf (770,829 bytes) |
| references |
Content-type: text/plain
|
| last changed |
2024/03/08 14:07 |
|
