Speaker
Description
Chinese traditional villages serve as crucial carriers for rural revitalization and cultural heritage preservation, with their tourism-driven revitalization emerging as a key approach to achieving regional sustainable development and safeguarding historical culture. The morphology and organization of street spaces in these villages not only embody rich regional cultural characteristics but also significantly influence tourists' emotional perceptions and behavioral patterns. Emotional perception, as an important metric for evaluating the spatial and cultural characteristics of villages and tourism experiences, holds great significance for optimizing the spatial design of traditional villages. However, existing studies primarily focus on static descriptions of spatial forms, overlooking the dynamic impacts of spatial elements on tourists' emotions. Furthermore, traditional methods such as surveys and interviews are often subjective, time-consuming, and inadequate for capturing the distribution of emotions and behaviors in real-world scenarios.To address these gaps, this study introduces deep learning techniques to construct a quantitative evaluation system for tourists' emotional perceptions, exploring the correlation mechanisms between street spatial elements and tourists’ emotional preferences in tourism-oriented traditional villages. A representative tourism-oriented traditional village was selected as the study area. Images of tourists’ facial expressions were captured at designated points across different streets in the village. These images were combined with a manually labeled dataset for emotion and age classification to train and optimize the YOLOv5 deep learning model. During data processing, techniques such as RGB value adjustment, image flipping, and augmentation were applied, achieving an accuracy of 0.81 on the validation dataset, thereby ensuring the precision and robustness of emotion recognition. The optimized model accurately identified four types of emotions (smile, surprise, calm, and negative) and three age groups (elderly, middle-aged, and young) among tourists, subsequently generating spatial distribution maps of emotional and age-based preferences.Using the spatial distribution data of emotions, 13 physical spatial elements related to street organization, morphology, and interface were selected for multiple regression analysis. The results indicate that street spatial separation is the most significant factor influencing tourists' emotions, with a regression coefficient of 0.184, identifying it as a critical determinant of positive emotional perception. Additionally, elements such as water systems, façade smoothness, and commercial diversity also significantly enhanced tourists' emotional perceptions. This study identified five key street elements that positively impact tourists' emotional experiences and proposed specific recommendations for optimizing street space design. These findings provide scientific guidance for enhancing tourism experiences and promoting the sustainable development of traditional villages.By innovatively applying deep learning techniques to the study of emotional perceptions in traditional villages, this research overcomes the subjectivity and limitations of traditional methods. It enriches the theoretical and practical understanding of the regional spatial characteristics of village streets, offering scientific pathways and novel methodological support for the integrated development of cultural tourism and rural revitalization.
References
[1]Zeng, C., Song, Y., He, Q. and Shen, F. (2018) 'Spatially explicit assessment on urban vitality: Case studies in Chicago and Wuhan', Sustainable Cities and Society, 40, pp. 296–306. Available at: https://doi.org/10.1016/j.scs.2018.04.021.
[2]Mouratidis, K. and Poortinga, W. (2020) 'Built environment, urban vitality and social cohesion: Do vibrant neighborhoods foster strong communities?', Landscape and Urban Planning, 204, Article 103951. Available at: https://doi.org/10.1016/j.landurbplan.2020.103951.
[3]Noji, S. and Kishimoto, T. (2020) 'Analysis of age, gender ratio, emotion, density of pedestrians in urban street space with image recognition: Case study mainly around Shibuya Station', Journal of the City Planning Institute of Japan, 55(3), October.
[4]Liu, Y., Li, Z., Tian, Y., Gao, B., Wang, S., Qi, Y., Zou, Z., Li, X. and Wang, R. (2024) 'A study on identifying the spatial characteristic factors of traditional streets based on visitor perception: Yuanjia Village, Shaanxi Province', Buildings, 14(6), p. 1815.
[5]Xu, G., Zhong, L., Wu, F., Zhang, Y. and Zhang, Z. (2022) 'Impacts of micro-scale built environment features on tourists’ walking behaviors in historic streets: Insights from Wudaoying Hutong, China', Buildings, 12, p. 2248.
| Keywords | Facial expression perception; Deep learning; Alley space; Tourist Emotions; |
|---|---|
| Best Congress Paper Award | Yes |
