Research And Implementation Of Terracotta Warrior Fragment Matching Technology Based On Fracture Surface

The Terracotta Warriors and Horses of Qin Shihuang is an outstanding representative of ancient Chinese pottery cultural relics and has extremely high research value.However,due to various natural or human factors,most of the cultural relics have been damaged to varying degrees when excavated.Using computer-assisted virtual restoration of cultural relics can effectively reduce the restoration period of cultural relics,avoid secondary destruction of cultural relics,and can solve many shortcomings of manual restoration in the past.Related research has become a research focus of computer graphics.Existing cultural relic fragment matching methods generally have the problems of single features and low efficiency.For the cultural relic fragments that are missing a lot of details on the fracture surface and the fracture site is seriously damaged,the matching effect in practice is not ideal.This article takes such terracotta relic fragments as the research object,and improves the matching effect from both accuracy and efficiency.The research content is as follows:(1)A multi-feature constraint based fracture surface matching method is proposed,which combines the thickness characteristics of fragment fracture surface,surface geometric texture features and fracture profile curve features to construct multi-layer constraints to complete the fragment matching process.First,the thickness characteristics of the fracture surface of cultural relic fragments are extracted,then complete the initial matching of the fragments based on the thickness histogram.Then define the fracture contour matching constraints based on whether the Euclidean distance between the feature points on the fracture contour line and the included angle are similar;define the geometric texture adjacency constraints based on whether the geometric texture line of the fragment surface can be connected,and then build a corresponding matching algorithm to complete the fragment matching,Finally calculate the fragment rigid body transformation parameters and align precisely based on the ICP algorithm.According to the experimental results,it can be proved that this method can solve the problem of low accuracy of single feature matching and improve the accuracy of fragment matching.(2)A fracture surface matching method based on contour curve is proposed,and feature vectors are constructed by combining corner information,and a two-level matchingstrategy is used for fragment matching.First,the fracture surface is divided and the contour line is extracted,and then the corner point information of the contour line is combined to construct the multi-feature vector at the corner point,and the matching function and matching algorithm are constructed to complete the rough matching of the fragments.Then extract several sampling points on the curve segment of the adjacent corner points,calculate the Euclidean distance from the sampling point to the chord length and construct the arc chord distance sequence,and use the longest common subsequence algorithm to complete the fine matching.Finally,the transformation matrix is obtained by the quaternion method,and the accurate alignment is based on the ICP algorithm.According to the experimental results,it can be proved that this method can effectively improve the matching efficiency based on the correct matching of fragments.(3)Designed and implemented a virtual restoration system for terracotta debris.The effectiveness of the matching method in this paper is verified and applied to the restoration process of terracotta fragments,which provides guidance for the merging of solid fragments.