Image Analysis of Industrial Heritage Based on Web Photo Data: A Case Study of Shanghai International Fashion Center

In the context of urban renewal, how the historical information and spatial culture of industrial heritage in transition are disseminated and perceived by the public needs to be emphasized. Urban imagery research based on big data has become an important trend, but due to the limitations of data and methodology, there is a relative lack of research on micro-scale such as industrial heritage sites. The paper took Shanghai International Fashion Center (SIFC) as the research object, classified 12 578 pictures related to the built environment collected on social media, and batch restored and tagged the photographer’s points based on 3D model combined with manual correction, and then analyzed the characteristics of imagery composition and their seasonal variations, spatial structural features and imagery in terms of public activities, public space, landmarks and natural landscape in the SIFC, and revealed the characteristics of spatial structure and imagery evolution of the SIFC. The study found that: 1) visitors’ perceptions of the renovated park are mainly reflected in the public space and public life, followed by the perception of the industrial heritage; 2) the main spatial structure of the original industrial heritage is basically retained after redevelopment, but some elements, such as the streets, have received insufficient attention; 3) the commercial art elements have weakened people’s perception of the industrial cultural elements, while giving the place vitality again. The paper revealed the bottom-up subjective perception characteristics, which can provide a reference for further understanding the value dissemination law of industrial heritage and the creation of regional characteristics.