| Forests are important and indispensable resources for the biosphere,providing a rich material basis for human survival and development.How to get to know forest resources correctly,manage forest resources scientifically and furthermore,realize the sustainable use of forest resources has always been the hot spot and focus of researchers.Forest resources inventory is an important means to obtain forest resources information.The efficient,accurate and convenient acquisition of the stand description factor is of practical significance to the correct grasp and scientific use of forest resources.Therefore,In order to satisfy the different needs of forest inventory,this paper focuses on the survey at forest farm scale,including three aspects,namely,ground forest resources inventory,UAV(Unmanned Aerial Vehicle)aerial forest resources survey and air-ground integrated biomass measurement.It provides accurate,efficient and low-cost observation equipment and a complete set of technical methods for obtaining forest resources information.The main research results are as follows.(1)Ground forest resource inventory technologyIn order to accurately mark the corner points of the sample sites in the ground forest resource inventory,a forest positioning device was designed and implemented based on Ultra Wide Band(UWB)relative positioning technology and smart-phone absolute positioning technology,and the forest positioning accuracy of the device was verified.The results show that the root mean square error(RMSE)and Bias were 0.05～0.11m and 0.02～0.01m,respectively.In order to realize the integration of sample plot deployment and stand parameter measurement,the UWB Intelligent tree-detector was designed and developed based on UWB positioning technology.The device integrates hardware modules such as UWB positioning module,angle sensor,CCD lens,etc.There are two types of software system:general observation mode of plot and three-dimensional observation mode of forest plot.In general observation mode of the plot,it realizes the calibration of the corner points of the plot and the measurement of the DBH,tree height and stand position.In three-dimensional observation mode of the stand,the observation method of"ring shot"was designed to extract the stand description factors of tree position,DBH,height under branch and height of tree.The experimental results show that in general observation mode,the evaluation indexes such as RMSE and Bias of DBH were0.9～1.0cm and 0.1～0.2cm,respectively,and RMSE and Bias of tree height were 0.3～0.6m and-0.1～0.2m,respectively;in three-dimensional observation mode,the evaluation indexes such as RMSE and Bias of DBH were 0.9～1.3cm and-0.4～0.6cm,respectively,and RMSE and Bias of tree height were 0.4～1.2m and-1.1～0.2m,respectively.Under the three-dimensional observation mode of forest stand,the UWB tree-detector has the ability to acquire stand image and generate dense three-dimensional point cloud efficiently,which can be applied to fine and three-dimensional plot observation work.(2)UAV aerial photographic survey technologyThe UAV with integrated GNSS module and inertial navigation module were used to obtains the image in the study area.The UAV image processing is accomplished by using auxiliary beam adjustment and image free processing method,generating digital orthographic model(DOM)and normalized point cloud data of the flight area.In order to test and verify the accuracy of the forests DTM generated by the UAV image data,the digital terrain model(DTM)generated from measured data of differential GNSS receiver is taken as the standard.The results show that the mean absolute error(MAE)of the forests DTM generated by UAV aerial image data is 0.26m.The factors of Single wood identification,tree height,canopy width and stand description factor were extracted by using controlled marker watershed method and regional growth method.Compared with the plot survey results based on UWB tree-detector,the regional growth method is a suitable method for single tree segmentation in the study area.The investigation factors,such as average tree height,average canopy width and plant number density,were calculated.The results show that the accuracy of the stand description factors could meet the requirements of forest resources inventory,indicating that the stand description factor extraction method based on UAV high-resolution images could meet the needs of forest resources survey in low and medium densities.At the same time,UAV aerial photographic survey also has the advantages of low cost,easy operation,short cycle,etc.,as a recommended forest inventory method at the scale of forest farm.(3)Forest Biomass Estimation with air-ground Integration technologyThe method of"micro-sample plot+UAV aerial photogrammetry"is designed as a method of forest biomass estimation with air-ground integration for forest farm scale,and the optimal model between the number of field monitoring plots and the reliability of forest resources inventory is determined.The biomass calculated from the plot survey data of UWB tree-detector was used as the dependent variable.The texture feature information extracted from DOM data and the canopy height variables extracted from normalized point cloud were used as the independent variables.And the Pearson’s correlation coefficient method was used to eliminate the independent variables with low correlation.A binary polynomial nonlinear model considering the characteristic relations among independent variables was proposed and compared with three biomass calculation models,namely,multiple linear regression,support vector regression and BP neural network.The results show that theR~2 of modeling precision and validation accuracy of binary polynomial nonlinear model is both above 0.68,the overall fitting precision is better than the other three models.The model has completed the forest biomass estimation and produced a spatial distribution map of biomass in the key camping area of the forest farm,and provided basic data for the preparation of the new management plan of the grassland forest.To sum up,this paper takes the forest resources inventory at forest farm scale as the research content.The research and development of the equipment for sample plot survey can realize the forest survey of the plots.The research and development of stand description factor inversion technique of aerial photogrammetry can achieve forest resources inventory at forest farm scale.The technology and method system of forest biomass estimation at the scale of forest farm was formed.The forest resource inventory of the integrated air-ground can make full use of the advantages of these two methods of monitoring forest resource.The technology and method system of forest biomass estimation at the scale of forest farm was formed. |
PDF Full Text Request