基于恢复性感知的城市滨水绿地景观要素对公众健康行为影响  

Impact of Urban Waterfront Green Space Landscape Elements on Public Health Behaviors Based on Restorative Perception

作　　者：吴逸 裘鸿菲[1,2] 罗心玥 胡亚萍 WU Yi;QIU Hongfei;LUO Xinyue;HU Yaping(College of Horticulture&Forestry Sciences of Huazhong Agricultural University;Key Laboratory of Urban Agriculture in Central China,Ministry of Agriculture and Rural Affairs)

机构地区：[1]华中农业大学园艺林学学院 [2]农业农村部华中都市农业重点实验室

出　　处：《风景园林》2025年第3期119-126,共8页Landscape Architecture

基　　金：国家自然科学基金面上项目“基于蓝绿协同的城市湖泊公园景观绩效与优化调控研究——以武汉市为例”(编号31770753);中央高校基本科研业务费专项资助项目(编号2662018PY087)。

摘　　要：【目的】城市滨水绿地内部空间景观要素与健康行为之间的复杂关联亟待深入剖析,恢复性感知作为自然环境中影响个体行为决策的关键因素,为探究二者关系提供了新视角。【方法】以武汉市沙湖公园93个景观节点为研究对象,结合计算机视觉技术、问卷调查和行为观测等方法,深入探究滨水绿地景观要素、恢复性感知对公众健康行为(恢复、体力及社交活动)的影响。【结果】自然景观有助于提升公众的恢复性感知,而高比例的人工要素则产生相反效果;不同健康行为存在空间分布差异且显著影响要素各不相同;恢复性感知与恢复活动、社交活动呈显著正相关。【结论】揭示了3类健康行为对节点景观要素的偏好差异,并证实了恢复性感知对促进健康行为的积极作用。未来可通过强化自然景观建设、合理规划空间色彩及实施节点差异化设计等措施,有效促进公众健康行为开展,进而提升滨水绿地健康效益。[Objective]Urban waterfront green spaces,as a typical representative of urban blue-green spaces,can serve as an important venue for the public to engage in various outdoor health-related activities.Although previous research has confirmed the close relationship between the diverse and complex landscape elements of waterfront green spaces and health behaviors,which mainly focuses on comparisons between overall waterfront green spaces,while seldom considering how landscape elements within these green spaces affect health behaviors.Meanwhile,restorative perception has been identified as a key factor influencing individual behavioral decisions in natural environments,which can provide a new perspective for exploring the relationship between the aforesaid landscape elements and health behaviors.Therefore,this research aims to delve into how landscape elements in urban waterfront green spaces influence public health behaviors from an internal spatial perspective,with restorative perception as a mediator.[Methods]This research takes Wuhan Shahu Park as an example and selects 93 landscape nodes within the park as the research objects.Initially,a DeepLabv3+semantic segmentation model tailored for landscapes in waterfront green spaces is developed through manual training.This is coupled with MATLAB-based color quantification and assignment statistical techniques to comprehensively and meticulously quantify landscape elements across six dimensions:Space,nature,artificiality,waterfront characteristic,color and entity.Subsequently,field surveys are conducted to gather public assessments of restorative perceptions(being away,fascination,extent and compatibility)at landscape nodes.Behavioral observations are also employed to document specific instances of public engagement in restorative,physical,and social activities within these sites.Ultimately,data analysis methods,including multiple regression analysis and mediation effect analysis,are applied to explore the interrelationships among landscape elements,restorative perceptio

关 键 词：风景园林 城市蓝绿空间 景观要素 景观感知 健康行为 计算机视觉 

分 类 号：TU986[建筑科学—城市规划与设计]