Design And Experimental Study Of UAV Image Forest Resources Inventory System

The rapid development of new science and technology such as uav technology,GIS,RS,Internet of things,cloud computing and big data has brought unprecedented development opportunities to the innovation of forestry resources investigation technology.Theoperation cycle of traditional forest survey often is long,so there is no doubt that the forest information update cannot follow the development needs of The Times,and the precision of the results is more affected by human factors.To meet the current forestry resources investigation for the efficiency and accuracy demand of information updates,fully understand the forestry industry demandand a lot of reading literature are presented in this paper.On all these bases a set of new forest resources investigation system is established.System can be used in forest resource survey data access and update in tiae,greatly shorten thetimeof traditional forestry survey Therefore,it provides technical support for the sustainable development of forestry resourcesThe system combines component GIS with C/S development mode;Complete system interface is designed with DevExpress Components;In the aspect of database designing,the operation speed of software is accelerated by means of information storage.Both the Oracle and spatial data engine ArcSDE are seamless connection,completing management system database;The software development platform is selected from Microsoft Visual Studio 2010,which is developed by Microsoft.The software realizes the spatial zoning and the area adjustment function of all levels of forest farm,forest class and sub?compartment based on uav image.Based on the DEM data of UAV images,terrain information such as slope,slope direction and altitude infonnation can be extracted to the database function of sub·compartments.The function of forest parameters such as average tree height,average DHB,plant tree density,accumulation volume,composition of tree species,biomass and other forest parameters can be obtained based on the data and image data of the foreign industry.The software also has GIS basic functions such as information query,factor editing,map browsing and mapping.As to validate the precision of the software,this study is based on the experimental field of the wangdian experiment in the chifengcity of Inner Mongolia,using the fixed-wing uav and the field research to get the drone aerial footage from the research area and the data from the ground.The results show that biases of the elevation information and slope information are 3.28m(0.28%)and-0.03°(-0.14%)respectively in the extraction of terrain informations.In which the Root Mean Squared Error of elevation information is 3.02 m(6.23%),while RMSE of slope information is 1.53°(6.27%).As for the extraction of vegetation information,the bias of the mean diameter of the sub-compartment is 0.04cm(0.25%),and the RMSE is 0.65cm(4.46%).The bias of the average tree height of stand is-0.17m(-1.37%)and the RMSE is 0.64m(5.11%).The results also show that the bias of stand density is 8.88plant ·hm-2(0.87%),and the RMSE is 4.75plant·hm-2(4.16%);the bias ofvolume of sub-compartment is 1.73 m3·hm-2(1.61%),,andthe RMSE is 6.36 m3·hm-2(5.93%)?...