| 摘要: |
| 找出影响街道步行环境夜间安全感的关键因素是提
升街道夜间活力的重要前提，确定采用基于生理传感器的现场
感知实验方法对关键影响因素进行识别。首先结合相关理论与
街道环境整体特征，建立环境自变量；再以可穿戴传感器收集
的皮肤电反应(GSR)为安全感测度基础，通过“唤醒-效价”
模型得出被试安全感因变量；而后建立夜间安全感的可视化地
图，从地图中选取产生不同安全感的环境样本，并以最优子集
回归的方法建立识别模型。基于45名被试的现场感知实验结
果表明，对夜间安全感产生影响的环境因素可以分为关键影
响、非关键影响与无影响因素。而本次实验得到的关键影响因
素为建筑内透光、绿视率和街道界面透明度，证明了夜间街道
中“自然监视”策略的重要性，同时给出基于提升“自然监
视”的街道夜景设计策略 |
| 关键词: 风景园林 街道步行环境 夜间安全感 感知实
验 环境影响因素 皮肤电反应 可穿戴设备 |
| DOI： |
|
| 基金项目: |
|
| Identification of Key Influences on the Perception of Safety at Night in Street Walking Environment: Walking Perception Experiments Based on Wearable Physiological Sensors |
| ZHU Meng,WANG Canxiang,CHEN Jinfu |
| Abstract: |
| Finding out the key factors affecting the night safety of
street walking environment is an important prerequisite for improving
the night vitality of street. The key influencing factors is determined
by field perception experiment based on physiological sensors. Firstly,
combined with the relevant theory and the overall characteristics of street
environment, the environmental independent variables are established.
Secondly, the skin electrical response (GSR) collected by the wearable
sensor is used as the basis of safety measurement. Then, through the
"arousal-valence" model, the dependent variable of the subject's sense
of security is obtained. Finally, a night security visualization map is
established, and environmental samples with different security are
selected from the map. The optimal subset regression method is used
to establish the recognition model. The results of field perception
experiments on 45 subjects showed that the environmental factors
affecting night safety could be divided into key influencing factors, nonkey influencing factors and non-influencing factors. However, the key
influencing factors obtained in this experiment are the light transmission
intensity of building, the greening view and the transparency of
street facade, which proves the importance of nighttime street natural
monitoring strategies and gives street nightscape design strategies based
on "natural monitoring" |
| Key words: landscape architecture street walking environment sense of
safety at night perception experiments key factor GSR wearable sensor |
