Speaker
Description
Public spaces have always played a central role in shaping vibrant urban experiences. As the primary venues for residents’ daily public activities, city streets offer multiple functions, including commuting, leisure, socializing, and exercise (Hassen & Kaufman, 2016). High-quality street environments not only improve residents’ health and quality of life but also directly affect urban livability and community vitality (Hornyák et al., 2023). At the same time, streets serve as important urban tourism attractions (Qi et al., 2022; Chen et al., 2022). By walking through city streets, tourists immerse themselves in local visual landscapes and cultural atmospheres, interacting with the environment and enhancing their travel experience . Especially in the post-pandemic era, as global tourism recovers and “urban walking” gains popularity, the quality of street environments has become a critical factor influencing tourist experiences (UNWTO, 2024; Wu, 2023). Given that most journeys are on foot, further investigation into perceived walkability of the street environment is urgently needed.Existing research on street environment perception has predominantly focused on residents, particularly specific groups such as the elderly or children, with relatively limited attention to differences between tourists and residents in street environment evaluations. Moreover, traditional methods (e.g., questionnaires and walking diaries) are often time-consuming and labor-intensive, making large-scale urban applications challenging.This study develops an innovative approach to more broadly and accurately evaluate how urban street environments influence walking perceptions among residents and tourists, while also analyzing the similarities and differences in their perceptions. Subsequently, we explore key factors driving these perceptual discrepancies and propose design strategies and guidelines to enhance urban street walkability.This evaluation method consists of three steps. First, we use OpenStreetMap road network data and ArcGIS to generate street sampling points and obtain corresponding street-view images, forming a comprehensive base dataset. Second, we recruit volunteers to provide walkability ratings for a subset of these images, creating a training dataset to feed a random forest model that predicts walkability scores for all street images in the study area, accompanied by a spatial pattern analysis. Third, by applying image semantic segmentation, we extract street elements and integrate them with built environment features and perceived walkability scores using elastic regression, thereby identifying crucial factors influencing the differences in perceptions between residents and tourists.Considering factors such as residential density and tourist numbers, we select the Ming City District of Xi’an as our empirical study area. Enclosed by ancient city walls, this district functions both as a tourist destination (receiving 211,000 visitors over five days during holiday periods) and as a residential area of approximately 306,000 permanent inhabitants (as of 2022). This research is part of an ongoing project, and a comprehensive review of the data is still in progress. We anticipate that the findings will reveal significant differences in residents’ and tourists’ perceptions of walking environments and elucidate key factors contributing to these variations. By focusing on these perceptual differences, this study aims to highlight the complexity of dynamic urban flows and provide valuable insights for urban planners and policymakers. Ultimately, we seek to design inclusive and walkable public spaces that simultaneously satisfy residents’ needs and enhance the tourist experience.
References
Hassen, N., & Kaufman, P. (2016). Examining the role of urban street design in enhancing community engagement: A literature review. Health & place, 41, 119-132.
2. Hornyák, S., Karancsi, Z., Korom, A., & Győri, F. (2023). What does a tourist see, or, an environmental-aesthetic evaluation of a street view in Szeged (Hungary). Turizam, 27(2), 113-135.
3. Chen, J., Wu, Z., & Lin, S. (2022). The influence of neighborhood quality on tourism in China: Using Baidu Street View pictures and deep learning techniques. Plos one, 17(11), e0276628.
4. Qi Z., Li J., He Z., & Yang X,J.(2024). Influence of Urban Street Landscape Color on Visitor's Emotional Perception——Study based on Street View Image. Journal of Geographic Information Science,26(02):514-529.
5. Wu, D. (2023). City walk in a gap day: potential and opportunities for tourism and leisure. Tourism Review.
6. United Nations World Tourism Organization (UNWTO).World Tourism Barometer.https://www.ungeneva.org/zh/news-media/news/2024/09/97699/guojiluyouqiangjinfusu-jiejinyiqingqianshuiping
| Keywords | Street view image; Semantic segmentation; Elastic network regression |
|---|---|
| Best Congress Paper Award | Yes |
