Uncovering Perceptual Differences in Street Greenery Based on Residents Different Accessibility of Travel Mode: A Case Study of the Central District of Chengdu
作者
Qicheng Ma,Jiaxin Zhang,Yunqin Li
出处
期刊:Frontiers in artificial intelligence and applications [IOS Press] 日期:2024-03-22
Urban street spaces serve as crucial public realms within cities. The focus on urban renewal, coupled with the utilization of multi-source data and the evolution of computational algorithms, has ushered in a novel methodological approach to investigate these spaces. Using the central district of Chengdu as a case study, this research leverages open-source road networks, street-view imagery, and remote sensing images. Through the implementation of deep learning semantic segmentation and remote sensing interpretation, we derive two distinct street greening measurements: a human’s-eye perspective and a god’s-eye perspective. Furthermore, space syntax analysis facilitates the extraction of both ‘To-movement’ and ‘Through-movement’ walking accessibility patterns. Subsequent bivariate local spatial autocorrelation analysis of the two greenery perceptions and the two travel patterns reveals the following findings: (1) There exists a spatial mismatch between overall street greening and street walking accessibility; (2) Streets with higher integration metrics currently have better greening, suggesting future greening initiatives should prioritize streets with higher choice metrics; (3) Central and peripheral urban areas exhibit heterogeneous street accessibility-greenery clustering patterns, indicating an inequitable distribution of public resources within the city. This study underscores the imperative for urban planners to embrace a ‘people-first’ approach and strive for a more equitable distribution of urban public resources in future urban renewals.