Visual aesthetic quality assessment of the streetscape from the perspective of landscape-perception coupling

Although significant progress has been made in both the description of landscape biophysical characteristics (LBC) and the assessment of public aesthetic perceptions (PAP) stimulated by these characteristics, there is still a lack of theoretical research and practical exploration in coupling the dual assessment factors of LBC and PAP. This study aims to propose an assessment method based on the coupled interaction of LBC and PAP. The Wudadao historic district was selected as the empirical study area. First, the assessment indicators describing the biophysical characteristics of the landscape were identified, and the indicators were quantified. Multi-source data, GIS technology, and deep learning algorithms were comprehensively utilized to optimize the indicator algorithm while enriching the indicator types. Then, a network research survey and the scenic beauty estimation (SBE) method were combined to quantify the public perception assessment results. Finally, a back-propagation neural network (BP NN) streetscape aesthetic quality assessment model based on the coupling of BCL and PAP was established. In addition, based on the quantitative data of LBC and PAP, the landscape characterization factors with a significant influence on the public aesthetic assessment were further clarified through factor analysis and eye-tracking experiments. They were, in order, the planning factor, the historical factor, the natural factor and the spatial factor. The results show that the proposed assessment method and technical path can effectively improve the scientific, operational, and precise assessment of streetscape visual aesthetic quality. The assessment results can be used as an important basis for landscape decision-making in the field of urban planning, prompting landscape changes to be oriented towards the construction of a harmonious relationship between humans and the Earth and the realization of the goal of sustainable urban development.