The Rhode Island Tree Locations dataset is a high-resolution vector product that maps individual trees to support improved management of tree resources and 3D visualization. Trees are delivered as points, polygons and idealized circular crown polygons. It is derived from processing color infrared imagery from 2014 and LIDAR data flown in 2011. LIDAR is a remote sensing technology that enables accurate measurement of elevation for different ground surface types based on the light pulses reflected from the earth.
The Rhode Island Tree Locations dataset supports numerous applications for urban planning, 3D visualzation and ecosystem analysis.
A Digital Surface Model (DSM) was created from first returns of the LIDAR point cloud data and bare earth elevations were removed from the DSM to derive an above ground surface model. Buildings and tree canopy were differentiated in the above ground surface model using color infrared imagery through an expert rule based classification system. Tree heights were interpolated using the highest LIDAR return within each 1 meter pixel that was classified as tree canopy to generate the Canopy Height Model (CHM). The CHM was analyzed to discover potential tree peaks through an iterative spatial filtering process. The tree peaks were used as markers for creating tree crown polygons through watershed segmentation. The potential tree crown shapes were further analyzed using shape and spatial context to refine the final tree polygons.
|Area||1,079 sq. miles|