Implement Of Contour Curve Matching Technology On Reconstruction Of Archaeological Fragments

To lots of cultural fragments, it would take lots of work and would cause the fragments damaged again when we reassemble the fragments by manual,so we need design the computer-aided fragments recovery system. In this system ,fragment automatic reassembly was reassembling the broken objects from a large number of irregular fragments by means of matching technique and computer assistance. There were some key technologies in this system,such as the description of the curves,the selection of feature quantity and the design of the matching algorithms.This paper researched on the irregular fragment shape matching and fragment reassembly. Mainly works are as follows:1.Research on the two-dimensional contour curve matching algorithm.In the algorithm the similarity of the debris mainly depends on the boundary.In the matching process,selecting the connection point as the feature quantity,because the connect point can not only effectively reflect the local characteristics of the point but also reflect the characteristics of neighboring points. Results show the accuracy and efficiency of this algorithm is high .2.The selection of the geometric characteristics of the discrete points is also a key technology in the matching algorithm.In the three-dimensional matching algorithm,selecting the curvature and torsion as the characteristic quantities. An improved method of convolution is presented in this paper.In multi-scale space,this paper presents a method of linear interpolation. Results show the improved algorithm can effectively calculate the discrete points's geometry feature.3.This paper proposes an improved algorithm based on Hausdorff distance.In the algorithm the type of feature points is classified according to its neighboring surface and the similarity of two contour is measured by the Hausdorff distance of the two contours.And the normal vector is utilized to verify the matchable probability of similar space curves.Experiment shows the algorithm not only improve the matching speed,but also take into account adjacent surface characteristics in matching process,which reducs the rate of error matching.