夜跑友好视角下城市街道环境视觉感知评价  

Evaluation on Visual Perception of Urban Street Environment from the Perspective of Being Friendly to Nighttime Running

作　　者：陈崇贤 刘欣宜 邱添 刘京一 CHEN Chongxian;LIU Xinyi;QIU Tian;LIU Jingyi(the College of Forestry and Landscape Architecture,South China Agricultural University)

机构地区：[1]华南农业大学林学与风景园林学院

出　　处：《风景园林》2024年第4期36-43,共8页Landscape Architecture

基　　金：国家自然科学基金资助项目“迈向健康城市环境:对后疫情时代城市公共空间规划、设计和管理的响应”(编号72111530208);广州市科技计划项目“城市街道景观适老健康效益智能化评测及优化研究”(编号202201010046)。

摘　　要：【目的】随着现代生活方式的转变,城市居民对夜跑活动的需求日益增加。已有跑步环境研究较多关注日间情景,城市街道夜间环境对跑者的影响还有待进一步探究。【方法】以广州越秀核心区为研究区域,利用计算机视觉和人机对抗评分平台评估不同类型街道对夜跑的安全感、恢复力、舒适度、吸引力及环境质量等方面视觉感知的影响,并运用空间自相关方法和地理加权回归模型研究各类街道环境要素与夜跑环境视觉感知之间的关系。【结果】研究发现,滨水街道和商业街可提升夜跑者的舒适度和恢复力,历史文化街道可增强夜跑的安全感和吸引力,而餐饮街和商住混合区街道较难激发正面跑步体验。不同街道环境要素对夜跑环境视觉感知的影响存在差异。【结论】考虑夜跑环境视觉感知评价对优化城市夜间跑步环境具有指导意义,未来可以通过区分街道类型,采取有针对性的规划设计措施来提升不同街道的夜跑体验质量。[Objective] As contemporary lifestyles continue to evolve, urban residents have witnessed a notable surge in the demand for nighttime running activities. However,it is imperative to acknowledge that the existing research on environmental perception predominantly focuses on daytime running scenario, leading to a constrained comprehension of how nighttime street environments influence the perceptions of runners. Given the distinctive disparities between daytime and nighttime running contexts, variables such as illumination, pedestrian density and accessibility of nighttime streetscape hold significant potential to shape runners' perceptions of safety, comfort, resilience, etc. Therefore, there exists a compelling necessity for a more comprehensive investigation into how diverse categories of street environments during nighttime influence the perceptions of urban residents interested in nighttime running activities.[Methods] This research aims to evaluate how nighttime street environments influence the visual perception of nighttime running environment by taking in account the six key indicators of safety, resilience, comfort, perceived accessibility,attractiveness, and quality of street environment. The research is conducted in the core area of Yuexiu District in Guangzhou. The computer vision technology is used to extract elements from nighttime street environments, and the visual perception of nighttime running environment is assessed using a combination of crowdsourced human-computer interaction evaluations and a residual neural network model, based on which a spatial autocorrelation analysis is performed.Furthermore, a geographically weighted regression model is applied to better understand the influence of various urban street environment factors on the visual perception of nighttime running environment.[Results] Different types of streets vary greatly in how nighttime runners perceive them. Waterfront streets and historical and cultural streets often display heightened levels of safety, resilience, and allure

关 键 词：风景园林 视觉感知 计算机视觉 城市街道 体力活动 夜间 地理加权回归(GWR) 

分 类 号：TU986[建筑科学—城市规划与设计]