Research On Fracture Surface Matching In Reverse Engineering

In the 21 st century,reverse engineering technology has not only become an effective means of digesting and absorbing foreign advanced technologies,but also playing an increasingly important role in product innovation and design innovation,industrial design,clinical medicine,cultural relics rapid restoration,3D modeling,etc.The essence of reverse engineering is to use 3D scanning equipment to get point cloud data which should be quickly and effectively deal with.Fracture surface matching is an important content in reverse engineering,which has extremely important research significance and application value.In this thesis,we focus on the matching of point cloud with large thickness fracture surfaces,and studied the key technologies in the matching process,including the segmentation of fracture surface,the feature extraction of fracture surface.The main research progress of this thesis are summarized as follows:(1)According to the features of the feature points on the boundary between surfaces,we calculates volume integral invariants of fragments' surface vertex under multiple scales to find sharp and raised vertices as the cut-off points to divide the point cloud into several sub-surfaces by the depth-first search algorithm.Finally,the fracture surfaces are identified and extracted according to the degree of surface roughness.(2)We extract the convex and concave regions on the fracture surface based on the mean curvature vector method.According to the normal vector of the points on the convex and concave regions to determine the main direction,we establish a single axis coordinate system,and project the points of region to the coordinate.According to the the height values of projection point on the axis,we determine the local extreme point of the feature region.(3)We complete the feature extraction of the fracture surfaces.We establish two types of feature libraries based on the local extreme points of the feature regions: one is based on the Euclidean distance between the extreme points.We use the exhaustive method to search the similarity of the Euclidean distances between the extreme points,to determine the matching pair of the fracture surfaces,and verify the matching pairs with the angle;the second is to search the extreme points of the k nearest neighborhood as the fracture surface feature area,and calculate the centroid of neighborhood.With the distance value of extreme point and centroid,we establish the feature library.Also we use the exhaustion method to search the fracture surface matching pair.(4)We have designed a point cloud processing system,and use the quaternion method to determine the rotation matrix and the translation matrix,which are used to rigid body transformation.With an example,we realize the matching of debris,and verify the feasibility of the algorithm.