The Processing And Application Of 3D Point Cloud Data Of Zokor’s Skull

The skull shape parameters of myospalax zokor can reflect its age composition and growth status.The traditional method of measuring the contour parameters of zokor skull is mainly manual measurement,but manual measurement with tape measure,vernier caliper,measuring tape and measuring stick has many problems,such as time-consuming measurement,low accuracy and large measurement error.In order to solve the above problems,the skull of the northeast Zokor was taken as the research object in this thesis,and 3d reconstruction and shape parameters measurement of zokor skull based on 3D laser scanning technology were studied.The main research contents of this thesis are as follows:(1)Aiming at the problem of unstable and inaccurate segmentation results when traditional region growth algorithm is used to segment the point cloud image of zokor skull with complex background,an improved region growth point cloud segmentation algorithm of zokor skull is proposed.Firstly,the point with the smallest curvature is selected as the seed point to reduce the total number of regions and make the segmentation result more stable.Secondly,according to the local characteristics of the point cloud,the growth criteria were determined for regional growth,and the points whose difference between normal Angle and curvature was less than a set threshold were grouped into one group to improve the accuracy of high point cloud segmentation.Finally,the segmentation results are optimized and the ideal segmentation results are obtained.(2)An improved ICP algorithm is proposed to solve the problems of high requirements for initial point cloud pose and slow search speed of traditional ICP algorithm,which is divided into two parts: initial registration and precise registration.Firstly,the Super4 PCS algorithm is introduced to complete the initial registration of two point clouds,which makes the two point clouds roughly coincide and lays a good foundation for accurate registration.In the precise registration part,the kd-tree structure is established to accelerate the search speed of the nearest point and improve the efficiency of accurate registration.Experimental results show that compared with the traditional ICP algorithm,the improved ICP algorithm has better registration accuracy and convergence speed,and the algorithm efficiency is significantly improved.(3)Due to the shape parameters such as nasal bone length,toothspace length and orbital width,it is difficult to accurately measure the zokor by two-dimensional image.So this thesis use of 3d laser scanner,obtain the mole rat skull point cloud data,after a point cloud denoising,segmentation,registration,such as processing,the final use of Crust algorithm to reconstruct the mole rat skull surface model,and then on the basis of using geometrical relationship between point and point to calculate the mole rat skull cranial length,length of nasal bone and tooth space shape parameters,such as,The experimental calculated values of the shape parameters of the zokor skull were compared with the actual measured values,and the error was small.At the same time,the age composition of myospalax myospalax was more accurately divided by the analysis of body weight and shape parameters.