| Reassembling of broken cultural relic is a necessary means to reproduce the original precious cultural heritage, is one of the important work of the archaeological process, Experts artificial reassembling process is inefficient and may cause secondary damage and other problems. With the rapid development of visualization technology and computer digital equipment, the application of computer-aided method for virtual reassembling cultural relics fragments has become a hot issue. Existing computer aided methods exist the problem that fracture surface matching is not accurate and it need human interaction. For the recovery needs of terracotta warriors which unearthed from excavation process, this paper proposes a fracture surface matching algorithm based on contour constraints. Using this matching algorithm improves the accuracy of fracture surface matching.Designing and achieving a terracotta warriors fragments 3d model matching recovery prototype system. The main research work includes:(1)Proposing a method of defining the feature contours of fragments model based on the main and secondary contour. For the problem of outer contour can not accurately describe the fracture characteristics, respectively defining the outer contour of model and the intersecting line of outer surface and the fracture as the main and secondary contour, and connecting with the main and secondary contour to build the feature contours of the fragments model. Experiments show that the feature contours can more accurately describe the contour characteristics relative to the outer contour.(2) Proposing a method for extracting secondary contour of cultural relics fragments 3D model. Firstly, the region expansion algorithm is adopted to segment the surface of model,then using the improved boundary removal method to merge the initial segmentation surface, and then, according to the roughness and area of the surface to identify the fracture. Finally, using surface scanning method for the secondary adjacency growth of fracture to obtain the feature point set of secondary contour. This method can quickly and accurately extract the secondary contour of models.(3) Using a feature contour line matching method based on a triple. Firstly, the curvature and torsion of all feature contours are calculated as the characteristic vector. and then construct the similarity matrix for each two curves, confirm a pair of feature contours that can be matched according to the number and length of similar curve segment. Then,using the unit triples method for all similar curve segment of this pair of feature contours to calculate the candidate matching transformation matrix relate to this pair of feature contours. This method can accurately identify the matching curves and all candidate matching transformation matrix.(4) Proposing a fracture matching algorithm constraint by curve matching. Firstly, calculating the Z-buffer of fragment model and fracture surfaces relative to the reference plane,and then, calculating surface permeability and fracture surface matching error for each set of matching transformation matrix to obtain the optimal matrix. This approach optimizes the matching results by contours.(5) Designing and achieving a terracotta warriors fragments matching recovery system. The system is simple to operate,and its interface is clear.The feature contour and fracture surface can be extracted accurately with this system. It realize to the matching recovery accurately for the Terra Cotta Warriors.This model matching method proposed by this paper is applied to virtual reassembling for the broken warriors get from third excavation of Qin Shihuang mausoleum Terracotta Warriors first Pit and K9901 pit. Comparing with the traditional method,this method can improve the accuracy and efficiency of the matching fragments effectively. |
PDF Full Text Request