Measuring the perceptual features of coastal streets: a case study in Qingdao, China

Abstract The coastal streets are the most attractive urban space, improving spatial quality and public perception of is an important work regeneration. study used machine learning semantic segmentation, GIS Semantic difference (SD) etc methods to obtain data perceptual evaluation in Qingdao. Each six features, imageability, enclosure, human scale, transparency, complexity nature, was taken as dependent variables corresponding physical features independent variables. regression models were established influence rules parameters on obtained. Meanwhile, based results evaluation, overall divided into three types, open streets, mixed biophilic streets. In all types nature significant feature due high greenness; lowest because low landscape diversity. research provided theoretical technical support for regeneration improvement