Study On Single Tree Structural Parameters Extraction And Wood Properties Prediction Of White Birch Based On TLS

Daxing'anling and Xiaoxing' anling areas are rich in forest resources and important forestry bases in China.However,due to the common disturbance of natural factors and human factors in the past the total stock and woodland area of the Daxing'anling and Xiaoxing'anling have been greatly reduced,the forestry resources have undergone tremendous changes,the forest biodiversity has decreased significantly,the forest structure is unreasonable and the forest stability is weakened,has led to the formation of large areas of secondary forests in the area.This thesis used TLS technology to predict and evaluate the wood properties of secondary forests.It has reference value for the analysis and evaluation of wood properties,and provides a new measure and method for its management,and then provides a theoretical basis for guiding the inducing transformation,restoration and reconstruction of secondary forests and the processing and utilization of wood.The main contents of this study are as follows:(1)Four Betula platyphylla secondary forest plots in the Daxing' anling and Xiaoxing'anling were scanned by using a 3D laser scanner and the threshold was set after scanning the trunk point cloud.We used Hough transform algorithm to extract the information of tree position and the diameter at breast height(DBH),and used the growth direction of the tree trunk to obtain the height and the crown width of the tree,and used regression analysis to fit the estimated value and the measured value to judge the accuracy of the algorithm.Finally,the volume of canopy was estimated by the methods of voxel simulation and traditional volume calculation,and the differences between the two algorithms were analyzed.The results showed that the average recognition rate of single tree in the four plots was 86.5%;the coefficient of determination of the DBH,tree height and crown amplitude of the four plots were 0.82,0.79 and 0.83,respectively;using the voxel simulation method and the traditional canopy volume calculation method,the average difference of the crown volume in the four plots was 35.6%,and the average the coefficient of determination of the four plots between those two algorithms was 0.96.The fitting accuracy is high,and the voxel estimation method is relatively smaller than the traditional calculation method for estimating canopy volume.(2)The physical mechanics and anatomical properties of white birch were studied.The axial variation of physico-mechanical properties and the radial variation of anatomical properties were observed,and the correlation was statistically analyzed between the properties.The results showed that from the axial change,the basic density,flexural strength and flexural modulus of wood gradually decreased with the increasing of trunk height;compression strength parallel to grain increased at 1-3m of the trunk,and then decreased;from the perspective of radial variation,the anatomical properties of the mature birch(fiber length,width,aspect ratio,fiber double wall thickness,wall cavity ratio,duct ratio,wood fiber ratio and wood ray ratio)changed little;from the correlation analysis,the correlation coefficient between wood density and compressive strength,flexural strength and flexural modulus of elasticity was 0.72,0.82,0.79,respectively.At the same time,there is a strong linear correlation between mechanical properties of wood and wood density,mechanical properties of wood and anatomical properties.(3)Single tree structure parameters extracted by terrestrial three-dimensional,established a prediction model between the structural parameters of the single tree and the properties of the wood(physical,mechanical and anatomical properties).The results showed that:Selected the best model from the five prediction models by goodness of fit,and the fitting effect of the mechanical properties of white birch was better.The correlation coefficients of the optimal prediction models for compressive strength,flexural strength and flexural modulus were 0.77,0.70,and 0.71,respectively.For the anatomical model,only the fiber length was 0.602,the basic density was 0.450,and the correlation coefficients of other anatomical prediction models were less than 0.4;In terms of model verification,comparing the data obtained from the prediction model and the testing data,the fitting accuracy of the above wood property model was verified.In summary,this thesis mainly used the point cloud data measured by 3D laser scanner to extract the single tree structure parameters of the secondary white birch forest by software and algorithm,measured the physical and mechanical and anatomical properties passed the wood properties experiment,and established a prediction model between them.The research results can provide a new research method for wood properties prediction,and have a good reference value for the analysis and evaluation of secondary forest.