-
摘要: 针对计算机辅助文物虚拟复原中由于破损文物断裂部位边缘受损而引起的轮廓线不能充分表示断裂面几何特征的问题,提出了一种基于断裂面拓扑特征的破碎文物自动拼接算法.首先,定义碎片模型顶点显著度指标函数,提取断裂面特征点,依据Morse-Smale复形理论构建并简化断裂面的几何拓扑图;然后,通过定义基准点与0值面,从而计算目标点的对应高度差值,将拓扑图中四边形曲面构造成为能完整表示断裂面几何特征的特征描述符,并根据凹凸互补性计算初始特征四边形匹配集的误差,筛选出最优匹配集;最后,采用四元组方法计算旋转、平移矩阵,利用穷举搜索法实现碎片的精确拼接.实验结果表明,该方法针对断裂部位边缘受损的破碎文物模型可获得较满意的拼接效果.
-
关键词:
- Morse-Smale复形 /
- 刚体变换 /
- 特征描述符 /
- 四元数 /
- 穷举搜索
Abstract: In order to address the problem that the traditional break-curves methods fail to reassemble fractured fragments with incompleteness in contours, a novel automatic reassembly method is proposed using topological feature of fracture surfaces in this paper. First, the Morse-Smale complex on the fractured surfaces of fragments is constructed with curvedness as the indicator function. Then, the tangent plane is calculated through height comparison among key points and the quadrilateral descriptor is obtained by computing height difference. After that, according to the correlation between adjacent regions, optimal quadrilateral descriptors are selected. The rigid transformation matrices that maximize the contact area between surfaces are obtained by quaternion method, such that two fragments can be precisely aligned based on optimal rigid motion through exhaustive search. Experimental results show that satisfactory performance can be achieved by several uses of the algorithm on the fragments of the terracotta warriors.1) 本文责任编委 刘成林 -
表 1 兵马俑碎片实验数据
Table 1 Experimental datas of the Terracotta Army fragments
编号 碎片数量 网格总数 特征四边形总数 匹配精度 G10-18 7 674 477 253 0.924 G10-26 2 38 602 32 0.929 G10-36 6 591 641 191 0.961 G10-22 5 451 750 103 0.975 G10-19 3 94 276 93 0.924 G10-23 3 12 147 84 0.896 表 2 本文算法运行时间
Table 2 Execute times of the proposed algorithm
编号 T1 (s) T2 (s) T3 (s) 总时间(s) G10-18 68.461 17.436 7.246 93.143 G10-36 53.534 16.990 6.128 76.652 G10-22 30.157 12.631 4.021 46.809 G10-26 7.065 3.518 0.981 11.564 G10-19 17.913 6.349 2.065 26.327 G10-23 19.254 7.521 2.731 29.506 -
[1] Cooper D B, Willis A, Andrews S, Baker J, Cao Y, Han D J, et al. Bayesian pot-assembly from fragments as problems in perceptual-grouping and geometric-learning. In: Proceedings of the 16th International Conference on Pattern Recognition (ICPR'02). Washington, D. C., USA: IEEE, 2002, 3: Article No. 30297 [2] 樊少荣, 茹少峰, 周明全, 耿国华.破碎刚体三角网格曲面模型的特征轮廓线提取方法.计算机辅助设计与图形学学报, 2005, 17(9):2003-2009 doi: 10.3321/j.issn:1003-9775.2005.09.018Fan Shao-Rong, Ru Shao-Feng, Zhou Ming-Quan, Geng Guo-Hua. A method of extracting feature contour from triangular mesh surface model of fractured solid. Journal of Computer-Aided Design & Computer Graphics, 2005, 17(9):2003-2009 doi: 10.3321/j.issn:1003-9775.2005.09.018 [3] Oxholm G, Nishino K. Reassembling thin artifacts of unknown geometry. In: Proceedings of the 12th International Conference on Virtual Reality, Archaeology and Cultural Heritage. Prato, Italy: ACM, 2011. 49-56 [4] Willis A R, Cooper D B. Bayesian assembly of 3D axially symmetric shapes from fragments. In: Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, D. C., USA: IEEE, 2004, 1: I-82-I-89 [5] Zhang K, Yu W Y, Manhein M, Waggenspack W, Li X. 3D fragment reassembly using integrated template guidance and fracture-region matching. In: Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV). Santiago, Chile: IEEE, 2015. 2138-2146 [6] Huang Q X, Flöry S, Gelfand N, Hofer M, Pottmann H. Reassembling fractured objects by geometric matching. ACM Transactions on Graphics, 2006, 25(3):569-578 doi: 10.1145/1141911 [7] Winkelbach S, Rilk M, Schönfelder C, Wahl F M. Fast random sample matching of 3D fragments. In: Proceedings of the 26th DAGM Symposium on Pattern Recognition. Tübingen, Germany: Springer, 2004. 129-136 [8] Papaioannou G, Karabassi E A, Theoharis T. Virtual archaeologist:assembling the past. Computer Graphics and Applications, 2001, 21(2):53-59 doi: 10.1109/38.909015 [9] 李姬俊男, 耿国华, 周明全, 康馨月.文物碎块虚拟拼接中的表面特征优化.计算机辅助设计与图形学学报, 2014, 26(12):2149-2154 http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201412007Li Ji-Jun-Nan, Geng Guo-Hua, Zhou Ming-Quan, Kang Xin-Yue. Surface feature optimization for virtual matching of relic fragments. Journal of Computer-Aided Design & Computer Graphics, 2014, 26(12):2149-2154 http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201412007 [10] Sahner J, Weber B, Prohaska S, Lamecker H. Extraction of feature lines on surface meshes based on discrete Morse theory. In: Proceedings of the 10th Joint Eurographics/IEEE-VGTC Conference on Visualization. Eindhoven, The Netherlands: IEEE, 2008. 735-742 [11] 邱彦杰, 周雄辉, 柳伟.基于Morse-Smale复形的三角网格特征线提取.上海交通大学学报, 2010, 44(8):1074-1078 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=QK201001992212Qiu Yan-Jie, Zhou Xiong-Hui, Liu Wei. Feature lines extraction from triangular mesh based on Morse-Smale complex. Journal of Shanghai Jiaotong University, 2010, 44(8):1074-1078 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=QK201001992212 [12] Shivashankar N, Natarajan V. Parallel computation of 3D Morse-Smale complexes. Computer Graphics Forum, 2012, 31(3):965-974 https://www.researchgate.net/publication/51866285_Parallel_Computation_of_3D_Morse-Smale_Complexes [13] Winkelbach S, Wahl F M. Pairwise matching of 3D fragments using cluster trees. International Journal of Computer Vision, 2008, 78(1):1-13 doi: 10.1007/s11263-007-0121-5 [14] 张春亢, 赵学胜, 王洪斌.采用Morse理论的小尺度地形特征提取方法.测绘科学技术学报, 2015, 32(3):266-270 doi: 10.3969/j.issn.1673-6338.2015.03.011Zhang Chun-Kang, Zhao Xue-Sheng, Wang Hong-Bin. Feature points extraction of small-scale terrain based on Morse theory. Journal of Geomatics Science and Technology, 2015, 32(3):266-270 doi: 10.3969/j.issn.1673-6338.2015.03.011 [15] Gyulassy A, Bhatia H, Bremer P T, Pascucci V. Computing accurate Morse-Smale complexes from gradient vector fields. Topological and Statistical Methods for Complex Data: Tackling Large-Scale, High-Dimensional, and Multivariate Data Spaces. Berlin Heidelberg, Germany: Springer, 2015. 205-218 [16] 张春亢. 基于Morse理论的三角网格特征提取及简化研究[博士学位论文], 中国矿业大学(北京), 中国, 2016. http://cdmd.cnki.com.cn/Article/CDMD-11413-1016062794.htmZhang Chun-Kang. Research on features extraction and simplification from triangle mesh based on Morse theory[Ph. D. dissertation], China University of Mining & Technology, Beijing, China, 2016. http://cdmd.cnki.com.cn/Article/CDMD-11413-1016062794.htm [17] Andreadis A, Gregor R, Sipiran I, Mavridis P, Papaioannou G, Schreck T. Fractured 3D object restoration and completion. In: Proceedings of ACM SIGGRAPH 2015 Posters. Los Angeles, California, USA: ACM, 2015. Article No. 74 [18] 江刚武, 王净, 张锐.基于单位四元数的绝对定向直接解法.测绘科学技术学报, 2007, 24(3):193-195, 199 doi: 10.3969/j.issn.1673-6338.2007.03.011Jiang Gang-Wu, Wang Jing, Zhang Rui. A close-form solution of absolute orientation using unit quaternions. Journal of Zhengzhou Institute of Surveying & Mapping, 2007, 24(3):193-195, 199 doi: 10.3969/j.issn.1673-6338.2007.03.011 [19] 李姗姗, 耿国华, 周明全, 李姬俊男.基于表面邻接约束的交互式文物碎片重组.计算机辅助设计与图形学学报, 2016, 28(6):924-931 doi: 10.3969/j.issn.1003-9775.2016.06.007Li Shan-Shan, Geng Guo-Hua, Zhou Ming-Quan, Li Ji-Jun-Nan. Interactive reassembly of fractured fragments based on surface adjacency constraint. Journal of Computer-Aided Design & Computer Graphics, 2016, 28(6):924-931 doi: 10.3969/j.issn.1003-9775.2016.06.007 [20] 李群辉, 张俊组, 耿国华, 周明全.以轮廓曲线为特征的断裂面匹配.西安交通大学学报, 2016, 50(9):105-110 http://d.old.wanfangdata.com.cn/Periodical/xajtdxxb201609017Li Qun-Hui, Zhang Jun-Zu, Geng Guo-Hua, Zhou Ming-Quan. Fracture surfaces matching based on contour curve. Journal of Xi'an JiaoTong University, 2016, 50(9):105-110 http://d.old.wanfangdata.com.cn/Periodical/xajtdxxb201609017 [21] 李群辉. 基于断裂面匹配的破碎刚体复原研究[博士学位论文], 西北大学, 中国, 2013. http://cdmd.cnki.com.cn/Article/CDMD-10697-1013253357.htmLi Qun-Hui. Research on fractured slid recovery based on bracture surfaces matching[Ph. D. dissertation], Northwest University, China, 2013. http://cdmd.cnki.com.cn/Article/CDMD-10697-1013253357.htm