Research On Correlation Between Spatial Characteristics Of Residential Streets And Walking Flow Based On Open Data

Street as the main space carrier of walking,the correlation between its spatial characteristics and walking behavior has become a hot topic.Sidewak is the most frequent place for urban residents to walk.It is of great significance to explore the spatial characteristics of typical residential street pedestrian flow.Based on the multisource open data,this study explores the correlation between the spatial characteristics of typical residential streets and the walking flow,explains the mechanism of its action,and clarifies the range of its appropriate value.The influence intensity difference of the significant influence factors of various street walking flow is discussed,and the optimization strategy of residential streets space feature is put forward.In this paper,residential streets in the central city of Beijing are taken as the research object.Based on the theoretical understanding of urban street spatial characteristics and walking behavior,a framework of research on the correlation between residential streets spatial characteristics and walking flow is proposed.First of all,based on the street spatial characteristics,the residential streets spatial characteristics are divided into three categories of first-level indicators,including spatial morphological features,organizational morphological features,functional facilities convenience;Based on the two dimensional functional classification of street,the residential streets is subdivided into traditional street type,mixed type unit type,closed tube type : 3 classes 6 typical types.The sources of the data are based on OSM map,combined with ArcGIS spatial analysis tool to quantify street scale and interface features,SegNET image analysis tool is used to intelligently cut Baidu street view map,and pixel analysis tool is used to quantify green view rate.The visual morphological features such as encirclement degree,sky visibility,motorization degree and so on;Spatial syntax is used to quantify the morphological features of the organization,and the features of convenience of quantization function and facilities of Baidu map POI are combined with ArcGIS spatial analysis tools.On the basis of OSM map,walking traffic data is quantified by STRAVA website and ArcGIS spatial analysis tool.Secondly,through the SPSS20.0 to verify that there are significant differences in the spatial characteristics of each type of street,and make clear the accuracy and necessity of residential streets classification.Thirdly,with the help of ArcGIS,this paper makes a descriptive statistical analysis of the spatial characteristics of various types of streets,which lays a data foundation for exploring the correlation between the spatial characteristics of sidewalk and walking flow.The data analysis is based on the SPSS20.0 data analysis platform.Firstly,the impact factors of various typical types of residential streets walking flow are identified by correlation analysis.Secondly,the influence mechanism of various typical residential streets spatial characteristics on walking flow is explained by univariate linear regression simulation,and the appropriate range is defined by scatter plot,and the differences are compared horizontally.Thirdly,based on the fitting degree of the model,the significant influence factors of walking flow were selected,and the difference of influence intensity was compared according to the standardization coefficient.Finally,taking specific streets of specific types as an example,the spatial feature optimization strategy is proposed.