Detection of streets trees using aerial photography of the city of Curitiba, Paraná

Abstract

Inventories of trees generate large volumes of data that needs to be stored and updated so that they do not become obsolete. Thus, the objective was to detect street trees using GIS to support planning and tracking through spatial registration of shading the streets. The detection of street trees was  conducted using aerial photographs in a GIS environment and compared with the amount of existing street trees provided by the conventional method  of inventory. Results indicated differences in the number of trees between the detection methods with the use of GIS in relation to the conventional  inventory. However, these differences were not statistically significant. It is concluded that the detection of trees in GIS can be used as a tool for quantification of trees in city streets. Thus, it may assist the conventional inventory by reducing the number of variables measured in the field and  consequently the high cost provided mainly by the displacement of staff.