Using crowdsourcing images to assess visual quality of urban landscapes: A case study of Xiamen Island

Abstract

Urban landscapes significantly affect human life and well-being. Visual factors are the most critical factors affecting environmental satisfaction and urban landscape quality. Previous studies have been focused on the single influence of landscape indicators on visual quality and ignored the relationship between indicators and landscape characteristics that may collectively contribute to visual quality. This study captures the public perception of urban landscape through social media image data and develops a new indicator model system based on structural equation modeling to assess the visual quality of urban landscape on Xiamen Island. Four visual characteristics (i.e., disturbance, complexity, naturalness, and artificial environment) are proposed to describe the composition of the urban landscape. The results show that: (1) Naturalness is the most important characteristic affecting the visual quality of the urban landscape, with a standardized factor loading of 0.82. (2) Artificial environment and complexity indirectly influence the visual quality of the urban landscape through naturalness and together form a coherent urban landscape. (3) Although disturbance has a direct effect on the visual quality of the urban landscape, most of the explained paths still affect the visual quality indirectly through naturalness. The results from this study highlight the importance of focusing on the relationship between the landscape indicators and characteristics and enhancing the understanding of urban landscape quality, with results that can potentially be used as a method for assessing future urban landscape changes and the effects of urban landscape policy decisions.