Font Size: a A A

Locate The Pith Position In Radial Direction By Using LiDAR

Posted on:2021-05-26Degree:MasterType:Thesis
Country:ChinaCandidate:D Y WangFull Text:PDF
GTID:2393330611969677Subject:Engineering
Abstract/Summary:PDF Full Text Request
The tree pith is one of the most important structures inside the trunk,which has a certain effect on the growth,performance and processing of the wood.Because pith positioning is the basis for forest tree measurement or trunk analysis,current researches focus on accurately locating it.At present,the direct measurement method with higher accuracy is mainly used to locate the pith,but these methods need to destroy the trunk and cause serious damage to the tree.In order to solve these problems,it is imperative to research and develop the non-destructive method to locate the tree pith.In this paper,a tree pith location system based on 2D laser LiDAR was developed after inferring and verifying the geometric property of pith.This system can obtain the point cloud data of the outer contour of the tree trunk through LiDAR under the premise of non-destructive.Experimental results show that this system can accurately mark a chord passing through the pith in the cross section of a trunk,which can fulfill the requirements for radial positioning of the pith when using growth cone or micro-drilling resistance drill.The main research contents of this article are as follows:(1)Proposing that the perimeter of outer contour of the cross section of the trunk longest chord passing through the pith can be evenly divided in allowable error range,and then verified it.Analyze the digital picture of the tree disc in Auto CAD.Manually determine the location of pith,then make the longest chord passing through the pith to divide the perimeter of outer contour of the cross section into two parts.Measure the value of the two parts of the circumference and calculate the error.Finally,it is verified that the error fluctuates around 5%.(2)Using 2D laser LiDAR to obtain the point cloud data of outer contour of the trunk with four fixed-point scans.After coordinate conversion,use the distance threshold to denoise and use Ant Colony Optimization(ACO)to sort the denoised data.Finally,the point cloud data is fitted and reconstructed by cubic spline interpolation.(3)Proposing a non-destructive radial positioning method for the tree pith.Extract the longest chord on the outer contour of the fitted trunk and verify whether it can bisect the outer circumference of the trunk cross section within the allowable range of error.If the chord passes the test,it can be deemed to pass through the pith,the the radial positioning of pith can be completed.The experiment found that for trees which cross-sectional shape approximates an ellipse,the error of this positioning method is only 1.46%.For trees with a trunk cross-sectional shape approximates to a circle,the pith can be directly located by mark out the longest chord on the cross-section without further inspection.Finally,by comparing with real sample trunks,it is proved that all marked chords can pass through the real tree pith,which verifies the accuracy of this positioning method.
Keywords/Search Tags:Pith, Pith position, Nondestructive testing, LiDAR, Point cloud
PDF Full Text Request
Related items