人视街景下的城市街道老年包容性昼夜评估OA

[Objective]This study aims to evaluate the spatial distribution characteristics of age inclusiveness of urban streets at daytime and nighttime and the influencing mechanisms,providing evidence to support the renewal and optimization of age-friendly streets.[Method]Taking Huadu District of Guangzhou as the study area,this research integrates human-perspective street-view images,subjective perceptional evaluations from the elderly,and deep learning techniques,to construct an age-inclusiveness prediction model from the four dimensions of safety,comfort,psychological friendliness,and social inclusiveness.Environmental visual elements are extracted using semantic segmentation of street-view imagery,spatial distribution patterns are analyzed under both daytime and nighttime scenarios,and spatial econometric models are employed to explore the spatially acting mechanisms underlying different dimensions.[Result]The age inclusiveness of streets in the study area is significantly lower at night than during the daytime,which is most pronounced in the dimensions of psychological friendliness and social inclusiveness,and spatial disparities become more evident at night.During the daytime,age inclusiveness exhibits the characteristic of significant spatial clustering(P<0.001),whereas the nighttime high-value areas shrink and become more dispersed.Daytime inclusiveness is primarily characterized by spatial spillover effects,while nighttime patterns are more strongly influenced by unobserved factors on the regional scale.[Conclusion]Age inclusiveness of the urban streets demonstrates significant diurnal differences and spatial heterogeneity,while the acting mechanisms are temporally differentiated.During the daytime,enhanced spatial quality may give play to spatial spillover effects,whereas nighttime conditions require strengthened lighting and safety measures to improve overall inclusiveness.Future research could incorporate longitudinal tracking data and multimodal physiological indicators to conduct multi-scale and multi-city comparative analyses,and integrate GPS data to further validate the universality and boundary conditions of day-night heterogeneity mechanisms.