一种有效的秦俑碎块匹配算法.PDFVIP

计 算 机 系 统 应 用 2017 年 第 26 卷 第 3 期 一种有效的秦俑碎块匹配算法① 赵夫群 1,2, 周明全 2 1(咸阳师范学院 教育科学学院, 咸阳 712000) 2(西北大学 信息科学与技术学院, 西安 710127) 摘 要: 针对秦俑碎块的三维网格数据模型, 提出了一种基于特征轮廓线的碎块断裂面匹配算法. 首先, 对数据 模型进行纹理贴图、去噪、补洞、简化数据模型等预处理, 然后提取碎块的主轮廓线和次轮廓线, 进而提取出碎 块的特征轮廓线, 最后根据角点对特征轮廓线进行分段, 并采用计算最长公共子序列的方法对分段曲线进行匹 配, 完成特征轮廓线的匹配, 从而实现碎块断裂面的匹配. 实验结果表明, 该算法是一种有效的、精确的秦俑碎 块匹配方法. 关键词: 秦俑; 碎块匹配; 特征轮廓线; 角点; 最长公共子序列; 曲率; 挠率 Effective Blocks Matching Algorithm of Terracotta Warrior ZHAO Fu-Qun1,2, ZHOU Ming-Quan2 1(School of Education Science, Xianyang Normal University, Xianyang 712000, China) 2(School of Information Science and Technology, Northwest University, Xi’an 710127, China) Abstract: Aiming at 3D mesh data model of Terracotta Warrior blocks, a block fracture surface matching algorithm based on feature contour is proposed in the paper. Firstly, the data model is pretreated, such as texture map, data denoising, hole filling, data model simplification and so on. Secondly, the primary contour and secondary contour are extracted, and then the feature contour is extracted too. Lastly, the feature contour is segmented according to the corner point in order to match the feature contour by the longest common subsequence method, then the contour feature matching is completed, and the fracture surface matching is achieved. The experimental results show that the algorithm proposed in the paper is an effective and accurate Terracotta Warrior blocks matching method. Key words: Terracotta Warrior; blocks matching; feature contour; corner point; the longest common subsequence; curvature; torsion 近年来, 图像配准技术发展迅速, 应用广泛[1-3], 线的碎块匹配, 很多研究者提出了一些匹配算法[6-9], 三维破碎刚体的匹配拼接也成为了一个研究热点. 破