摘要
研究古迹恢复中的三维模型准确重建问题。古迹恢复性保护修复过程中,需要对破损、磨损失色部位进行计算机三维重建,一旦破损部位较大或者磨损严重,残留的可恢复特征就会较少。传统的古迹视觉恢复方法在磨损严重或者破损较大的情况下,估计的恢复特征很难保证连贯性,导致重建效果不好。为此提出了一种采用多层次纹理映射估计模型的破损古迹重建方法。将多层次估计分析的相关原理与纹理映射关系模型结合,能够得到多层次纹理映射模型,将破损古迹特征点输入到该模型中,从而获取较为准确的特征映射关系。根据上述特征映射关系,建立破损古迹三维视觉模型,从而实现破损古迹的重建。实验结果表明,利用改进模型进行破损古迹重建,能够极大的提高重建古迹的真实度。
In this paper, the accurate model of three-dimensional reconstruction in monuments restoration was studied. This paper presented a method for damaged relics reconstruction based on multilevel texture mapping estima- tion model. Firstly, the related principle of multi-level estimation analysis was associated with the texture mapping relation model, and the multilevel texture mapping model was obtained. Then, the feature point of damaged relics was input to the model, thereby to obtain a more accurate characteristic mapping relation. Finally, according to the char- acteristics of mapping relation, the three-dimensional visual model of damaged relics was built, and the damaged monument was uhimately rebuilt. The experimental results show that the modified model used in the rebuild damaged monuments can greatly improve the fidelity of reconstruction monuments.
出处
《计算机仿真》
CSCD
北大核心
2014年第3期437-440,共4页
Computer Simulation
基金
浙江省工业设计科技项目(2013D40046)
关键词
三维视觉
古迹重建
纹理映射模型
3D vision
monuments reconstruction
Texture mapping model