Authors: Rui Si*, Harbin Institute of Technology(Shenzhen), Zuopeng Xiao, Harbin Institute of Technology(Shenzhen)
Keywords: Street view; Daytime vitality; machine learning; urban environment
Session Type: Paper
Start / End Time: 11:10 AM / 12:25 PM
Room: Governors Square 17, Sheraton, Concourse Level
Presentation File: No File Uploaded
Urban environment has been demonstrated to have associations with the time and space behavior of the active population. The era of big data provides researchers with a more sophisticated scale of research, more efficient and low-cost research data collection. Relevant research on the built environment in urban vitality mainly focuses on land use type and spatial layout, and pays less attention to eye-level street space, although streets are among the most popular venues for physical activities. Besides, the daytime vitality is the most representative part of the space vitality, so it is necessary to conduct an in-depth discussion of the daytime vitality of the block. This study takes Futian District of Shenzhen as an example to measure and identify the daytime vitality of urban blocks based on Baidu heat map data. Batch extraction of street space elements is done by using machine learning algorithms based on street view data. We used GIS to standardize the processing and spatial analysis of the two, and then used SPSS to analyze the correlation of the urban environment indicators of Street View identification. We delved into the relationship between the daytime vitality of urban blocks and the urban environment under the humanistic perspective and proposed an environmental optimization strategy to provide a basis for improving the vitality of the neighborhood.