Impact of Landscape Quality of Traditional Village Tourism Destinations on Spatial Vitality: an Empirical Study of Beigang Village in Pingtan County Based on Multi Source Data and Machine Learning
-
摘要: 乡村旅游地活力是传统村落旅游振兴的重要因素。近年来,多源数据与机器学习分析体系的广泛应用拓宽了乡村建成环境的研究视角,但目前乡村空间活力影响因素的研究中,缺乏对人主观感知要素等更细节性的成因探讨。以平潭县北港村为研究对象,首先将人口热力数据、POI(Point of Interest)数据、街景拍摄数据分类整理,其次用语义分割方法、公众感知评分获取景观要素占比与打分数据,最终使用XGBoost(eXtreme Gradient Boosting)方法构建模型并通过SHAP(Shapley Additive Explanations)方法解释其不同要素的贡献度。实验结果表明:1)蓝视率、开阔度、铺装度和脏乱度是对空间活力影响力较大的4类指标,同时具有一定的非线性关联,反映了特色景观建设对于乡村空间活力提升的重要意义。2)许多指标如绿视率、蓝视率呈现跳跃式聚集的分布模式,表明景观规划设计的连续性较弱。研究结论揭示了不同景观感知要素与空间活力的相互作用机制,拓展了多源数据与景观感知数据在中微观尺度空间中的应用,为乡村旅游地规划与景观设计提供参考。Abstract: The vitality of rural tourism destinations is an important factor in the revitalization of traditional village tourism. The widespread application of multi-source data and machine learning analysis systems has broadened the research perspective of rural built-up environments. However, in the current research on the factors affecting rural spatial vitality, there is a lack of more detailed exploration of the causes of human subjective perception factors. This study took Beigang Village, Pingtan County, as the research object. Firstly, the population thermal data, POI (Point of Interest) data, and street scene shooting data were classified and organized. Secondly, the semantic segmentation method and public perception score were used to obtain the proportion and scoring data of landscape elements. Finally, XGBoost (eXtreme Gradient Boosting) method was used to construct the model, and the contribution of different elements was explained through the SHAP (Shapley Additive Explanations) method. The experimental results showed that: 1) Blue vision rate, width, paving degree, and dirtiness are the four most influential indicators, and have a certain non-linear relationship, reflecting the important significance of characteristic landscape construction for enhancing rural spatial vitality. 2) Many indicators, such as green vision rate and blue vision rate, exhibit a distribution pattern of jumping aggregation, indicating weak continuity in landscape planning and design. The conclusion indicates that the fault characterization of different landscape elements and the interaction mechanism between landscape elements and spatial vitality have expanded the application methods of multi-source data and detailed landscape data in meso and micro scale spaces, providing reference for rural tourism destination planning and landscape design.
-
[1] SU M M,WALL G,WANG Y,et al. Livelihood sustainability in a rural tourism destination-Hetu Town,Anhui Province,China[J]. Tourism Management,2019,71:272-281. [2] 朱怡帆,何丹,范振杰. 乡村旅游地居民幸福感影响因素及其作用路径[J]. 地理科学,2023,43(9):1598-1607. [3] 简·雅各布斯. 美国大城市的死与生[M]. 南京:译林出版社,1961. [4] 扬·盖尔. 交往与空间[M]. 北京:中国建筑工业出版社,1992. [5] 冯月,余翩翩. 社区街道活力评价及影响因子辨析:以成都市为例[J]. 西部人居环境学刊,2019,34(6):18-24. [6] 丁杰,沈新. 旅游型乡村空间活力的分布特征及其影响机制:基于多源数据的宏村实证分析[J/OL]. 旅游科学:1-16[ 2023-07-25]. https://doi.org/10.16323/j.cnki.lykx.20230330.001. [7] 肖佳琦,宣蔚,孔子然,等. 基于多源数据的商业空间活力关联机制研究[J]. 工业建筑,2023,53(6):59-65. [8] LOPES M N,CAMANHO A S. Public green space use and consequences on urban vitality:an Assessment of European Cities[J]. Social Indicators Research,2013,113(3):751-767. [9] FRUIN J. Pedestrian planning and design metropolitan association of urban designers and environmental planner[J]. 1971. [10] 付鹏,肖竞,赵之齐,等. 基于机器学习的乡村聚落“空间—动力”耦合机制解析方法研究:以江苏溧阳市为例[J]. 西部人居环境学刊,2022,37(4):1-9. [11] 汪成刚,王波,王琪智,等. 城市活力与建成环境的非线性关系和阈值效应研究:以广州市中心城区为例[J]. 地理科学进展,2023,42(1):79-88. [12] WU J,LU Y,GAO H,et al. Cultivating historical heritage area vitality using urban morphology approach based on big data and machine learning[J]. Computers,Environment and Urban Systems,2022,91,101716. [13] 陈健平. 社区营造视角下乡村旅游高质量发展路径研究:以福建平潭北港村为例[J]. 安徽农业大学学报(社会科学版),2019,28(3):1-8. [14] 张丽英,裴韬,陈宜金,等. 基于街景图像的城市环境评价研究综述[J]. 地球信息科学学报,2019,21(1):46-58. [15] 黄竞雄,梁嘉祺,杨盟盛,等. 基于街景图像的旅游地街道空间视觉品质评价方法[J/OL]. 地球信息科学学报:1-15[ 2023-07-25]. http://kns.cnki.net/kcms/detail/11.5809.p.20221219.0958.002.html. [16] ZHANG P,ZHOU J,ZHANG T. Quantifying and visualizing jobs-housing balance with big data:a case study of Shanghai[J]. Cities,2017,66:10-22. [17] 吴志强,叶锺楠. 基于百度地图热力图的城市空间结构研究:以上海中心城区为例[J]. 城市规划,2016,40(4):33-40. [18] 叶怡君. 景区依托型传统村落公共空间活力评价与活化研究[D]. 桂林:桂林理工大学,2022. [19] SUN K,ZHAO Y,JIANG B,et al. High-resolution representations for labeling pixels and regions[J]. 2019. DOI: 10.48550/arXiv.1904.04514. [20] WU C,YE Y,GAO F Z,et al. Using street view images to examine the association between human perceptions of locale and urban vitality in Shenzhen,China[J]. Sustainable Cities and Society,2023,88,104291. [21] 韩君伟,董靓. 街道景观视觉评价量化指标及其有效性分析[J]. 西南交通大学学报,2015,50(4):764-769. [22] 窦燕,孔令富,王柳锋. 基于视觉熵的视觉注意计算模型[J]. 光学学报,2009,29(9):2511-2515. [23] SILVERMAN B W. Density Estimation for Statistics and Data Analysis[M]. Boca Raton:CRC Press,1986. [24] CHEN T,GUESTRIN C. XGBoost:a scalable tree boosting system[C]// Knowledge Discovery and Data Mining. ACM,2016. DOI: 10.1145/2939672.2939785. [25] LUNDBERG S,LEE S I. A unified approach to interpreting model predictions[J]. 2017. DOI: 10.48550/arXiv.1705.07874. -
点击查看大图
计量
- 文章访问数: 26
- HTML全文浏览量: 6
- PDF下载量: 1
- 被引次数: 0
登录
注册
E-alert
登录
注册
E-alert
下载: