Tree height extraction using high-resolution imagery acquired from an unmanned aerial vehicle (UAV)

Unmanned aerial vehicle (UAV) remote sensing technology plays an important role in the acquisition of tree parameters. In order to explore the feasibility of using high-resolution imagery of UAV to extract tree height, the Qiuji Coal Mine Forest Park of Shandong Province of eastern China was selected as the study area in this paper, the Pix4D software was used to process high-resolution imagery collected by UAV to generate the orthomosaic and 3D point cloud. The OTSU method was used to divide the 3D point cloud into two parts: the tree point cloud and the ground point cloud, then the top height of tree and the mean height of the ground were extracted according to the two parts of the point cloud. And the tree height was computed when the average ground height was taken as the height of tree root. The results showed that the tree point cloud and the ground point cloud could be accurately divided by OTSU method. And it was feasible that high resolution UAV imagery was used to extract the tree height, the absolute error of tree height was less than 80cm, the relative error was 16.2%, and the mean square error (RMSE) was 36.3cm. At the same time, the measurement of tree height was affected by the shape of crown. The height RMSE of Platanus acerifolia with broadly ovate crown was 29.2cm, the height RMSE of Sabina chinensis with conical crown was 50.9cm, the coefficient of determination R² of the measured height and the real height of the two kinds of trees were respectively 0.9920 and 0.8894, the accuracy of measured height of the Platanus acerifolia was significantly higher than that of the Sabina chinensis.