Associations between the quality of street space and the attributes of the built environment using large volumes of street view pictures

In this study, we focus on the quality of street space which has attracted high attentions. We discover associations between the quality of street space and built environment attributes through an ordered logistic model using massive street view pictures (SVPs) and data on street location, form, function and attributes. Before ascertain which built environment factors influence the quality of street space, we checked the concordance of the experts’ scores, as well as correlations between different dimensions through Kappa analysis and drew the distribution map of street space quality.We found that the value of intersection over union is 85.61% for scoring the street space quality by different people. The spatial quality of more than 75% streets are in the middle level with no obvious polarisation observed in the central area of Qingdao. In addition, for street quality index, all variables are statistically significant. The sequence is as follows: near-line rate > D/H ratio > slope > length of street > distance to administrative center > POIs diversity. The D/H ratio, near-line rate, slope length of street, distance to administrative center and POIs diversity have various associations on every dimension of street quality. They can prove useful for drafting more appropriate policy measures aimed at improving street quality.