Study on the Street Space in Historical Urban Areas under the Multidimensional Measurement of Crowd Perception:Taking Xiguan Region in Guangzhou as an Example
LIN Zezhao
LI Minzhi
Abstract：From the perspective of humanism,urban design increasingly emphasizes the experience of crowds in space.The key to assessing experience is to understand a crowd's perception of the environment,including their subjective perception of the built environment,natural environment,and urban services.In recent years,social media data and streetscape images have become important tools for studying crowds'perceptions in the urban environment.In the face of the spatial resource mismatch that widely exists in urban design,crowd perception data can reveal the difference between the real needs of the crowd and the urban environment,thus providing a scientific basis for optimizing urban design.
This paper uses computer vision and natural language processing techniques to interpret multiple types of data in the street space of Xiguan region in Guangzhou.First,the street element percentage and functional characteristics are quantified,and the spatial quality is scored.Then,semantic and emotional data in social media are analyzed to construct an imagery heat map of crowd perception,identify the elements of crowd concern and emotional tendency,and summarize the crowd's subjective perception characteristics.Finally,the collected data are analyzed in ArcGIS to compare the crowd perception data with the street spatial data.A comparison of the imagery heat network and the spatial subjective and objective evaluations uncovers the differences between the crowd behaviors and the physical spatial data so the streets with this feature can be extracted in bulk from the Xiguan region.
The study shows a significant difference between the scores of streetscape images and the crowd's actual emotional perception at the street scale.In addition,many street spaces with excellent spatial scores failed to attract attention from the crowd and lacked vitality.At the district scale,there are both overlaps and differences between the spatial structure perceived by the crowd and the existing layout of the street space.This suggests that the traditional urban design approach that relies only on the physical environment may not be able to fully meet the actual needs of the crowd.
The quantitative model of crowd perception established by multi-source big data can accurately capture the subjective feelings of the crowd and help the planning and design of street space in historical urban areas to be closer to the needs of actual users.The study explains the contradiction between crowd perception and physical spatial environment:high-quality physical spaces do not necessarily lead to a good user experience,and spaces with poor environments may show high vitality due to their unique historical and cultural backgrounds.Although there are deficiencies in material conditions,a large number of spaces in the Xiguan region have great development potential due to the attraction of history and culture.Planners should take these potential resources into full consideration and organically integrate historical streets with modern urban spaces to stimulate the diversified vitality of the streets.
Thus,this study proposes a street space optimization method based on multi-source big data through the quantitative analysis of crowd perception data and emphasizes the importance of the"human-oriented"design concept in the protection and development of street space in historical urban areas.Future studies can further expand the dataset and model to improve the prediction accuracy of spatial perception to gain a deeper understanding of the complex relationship between crowd perception and urban space.
Keywords：street space in historical urban areascrowd perceptionstreetscape imagesocial mediaXiguan region in Guangzhou
Publication Date：2024-11-28
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：9( 41-49 )