Spatial differentiation characteristics and influencing factors of the green view index in urban areas based on street view images: A case study of Futian District, Shenzhen, China

Urban green space provides a variety of benefits, and the green view index (GVI) is regarded as an effective indicator to evaluate the quality of green spaces from a human perspective. To investigate the drivers of the spatial differentiation in GVI in urban areas, the GVI was calculated based on Baidu street view images and semantic segmentation in Futian District, Shenzhen, and its spatial variation characteristics were explored. The dominant influencing factors of GVI were analyzed and screened by applying the Pearson correlation coefficient and optimal parameters-based geographical detector (OPGD). The results showed that the overall GVI was high in the study area, with an average value of approximately 28%. The scale effect of GVI was not obvious, but the spatial heterogeneity was distinct, and the hot spots and cold spots of GVI had an evident clustered distribution. The spatial variation of GVI was influenced by the vegetation cover, landscape pattern, built environment and socioeconomic factors. The normalized difference vegetation index (NDVI) was the most dominant factor affecting the spatial differentiation of GVI, followed by land cover, fractal dimension index, patch area, building density and proximity. Furthermore, the interaction of the influencing factors had a higher degree of explanation than a single factor. In highly urbanized areas, exploring the factors affecting the spatial differentiation of GVI can provide a basis for optimizing the layout and structure of green spaces, contributing to a better perception and quality of urban greening.