Study On The Method Of Forest Information Extraction And Interpretation By Model Aircraft

Forest resource survey is in order to grasp the interrelation between dynamic changes of forest resources and the relationship between natural environment,as well as providing basic data support for forestry zoning,design,and management.With the development of remote sensing and photogrammetry technology,UAV photogrammetry technology has been widely used in it.In this study,we take the small model aircraft UAV as the video data acquisition platform,mainly focus on small-scale,high-precision forest resources survey,the research content is as follows:(1)High-precision digital surface model and digital orthophoto map production technology.The Pix4D mapper software was used as the aerial image processing tool to process the original image with GPS data,digital orthophoto DOM and the digital surface model DSM.(2)The Object oriented image segmentation technique was used to extract the canopy and canopy density parameters.Using object oriented image segmentation technique and choose appropriate segmentation scale parameters to extract the information of crown and crown density parameters,to compare with the method of manually plotting the vector graph.The results showed that the average precision of the average crown was 0.85,and the average precision of the forest canopy density extraction was higher than 0.90.(3)Using remote photogrammetry instrument to remote remeasure complex conditions,unreachable wood in study area.Remote photographic tree survey system,which is made up with the smart phone and the long-distance telescope,complete the observation of the target tree,realizing the supplement to the ground investigation scheme.The experimental results show that the accuracy of tree height is 92.62%and the measurement accuracy of DBH is 95.16%,which can meet the requirements of forestry investigation.(4)The inversion model of broad-leaved forest and coniferous forest in the study area was established and the optimal model was selected.DBH-crown model,DBH-crown/tree height model and volume inversion model of broad-leaved forest and coniferous forest were established by using the data of crown and canopy density parameters extracted from UAV images combined with ground sample survey data in the study area.In summary,this study mainly focus on the small and medium-sized model aircraft UAV for the forest resources survey,forming the UAV+ground sample observation+remote photography observation as one of the forest observation system,With a view to improving the efficiency of forest resource surveys and improving forest survey programs,as well as providing new directions for forestry survey.