Visibility Assessment Of Urban Street-level Greenery Using Street View Imagery And LiDAR

Street-level urban greenery is artificial network that is growing between the sidewalk and the curb.As a physical components of urban ecosystems,street tree not only provides many functions in urban context that benefits people's quality of life,such as energy saving,provision of shade,and aesthetic values,but also plays a key role in shaping urban residential landscape.At the heart of city structure lies the urban road networks that links both the nature of the built environment.City streets become desirable open space for residents having a daily contact with nature.With the expansion of urban road system,people's viewing of street greenery as a sensory function has been identified related to the attractiveness and walkability of residential streets.Developing a greenway network that connects natural settings and residents has become a significant concern.Therefore,increasing people's esthetic rating of urban scenes by optimizing the distribution of the distance and accessibility of green areas,lays the foundations for the success of a greening program.Today,Google Street View(GSV)panorama lets people experience the streets and makes you feel like you're standing right there at that very spot.As a ‘big data' that including visual information about streets,the GSV images has been proved as an effective data source for design and planning purposes.In recent years,Li DAR technology becomes a powerful tool to detect irregularly shaped 3D objects,as it can precisely generate the shape,geometry and position of targets automatically with high spatial resolution.Applying these date sources in studying the visibility of street greenery provides a new insight for assessing the aesthetic attributes of urban scenes.In the face of strong demand of residential livability,the visibility of street trees is assessed in terms of the visual perception of pedestrians.In this study:(1)Li DAR data and GSV images were used to automatically extract canopy information based on object-oriented classification and three-dimensional modelling method;(2)A new Green View Index(GVI)was proposed as a measure of mapping the visibility of street greenery based on the extracted data and the proximity and adjacency of pedestrian to trees;(3)the relationship between the GVI distribution and different urban scenes is analyzed for validation purposes.Results showed that the GVI models could be used as an objective tool to represent the real amount of greenness that can be seen by a pedestrian on the ground.Thus,the proposed model is expected to help planning directors to better understand the sensory functions of street greenery,make the workflows more efficient,and generate valuable spatial reference for future urban greening programs.