摘要
为对秦始皇陵出土的彩绘铜车马采取非接触且科学可靠的方法开展研究,选择使用高光谱成像技术,以非接触方式无损地获取文物表面信息。通过对信息的处理与分析,实现对于文物表面彩绘纹样的发现与复原工作。针对秦始皇帝陵出土的二号青铜车马彩绘纹样的受损情况,提出了结合高光谱成像技术与计算机图像技术的数字复原方法。主要是利用了高光谱成像技术“图谱合一”的优势:一方面利用其在近红外区域的高光谱成像结果检测并复原出彩绘纹样难以识别的图案边缘信息;另一方面利用高光谱数据中的光谱信息,将铜车马彩绘纹样上未知矿物颜料的光谱同实验建立的矿物颜料标准光谱库进行拟合,识别出铜车马上彩绘所采用的颜料种类(雄黄1、石青5、石绿1、黄丹、铅白等),从而获得色彩信息。实验结果显示该方法可以得到较为完整的铜车马彩绘纹样的数字复原图像,并为今后对于铜车马更深入的研究和应用提供有利支撑。
Hyperspectral imaging can acquire information on the surfaces of cultural relics in a contactless and nondestructive way.Processing and analysis of the information is very helpful for the discovery and restoration of painted patterns on the surfaces of cultural relics.Based on damage to the painted patterns on the No.2 Qin bronze chariot unearthed from the Emperor Qinshihuang’s Mausoleum Site,we propose a digital restoration method combining hyperspectral imaging and computer image technology.On one hand,the near-infrared hyperspectral image was used to detect and recover the edges of the patterns which are difficult to recognize;on the other hand,the spectra of the unknown pigments on the painted patterns were matched with those in an established standard library of mineral pigments to identify the type of pigments(realgar 1,azurite 5,malachite 1,yellow lead and lead white,etc.)used in the painting of the bronze chariot.The results show that the method can obtain a complete digital restoration image of the painted patterns,which can provide favorable support for further research on the bronze chariot and its application.
作者
马琳娜
马生涛
张加万
韩冬
MA Linna;MA Shengtao;ZHANG Jiawan;HAN Dong(School of Software,School of Intelligence and Calculation,Tianjin University,Tianjin 300350,China;Emperor Qinshihuang’s Mausoleum Site Museum,Xi’an 710600,China)
出处
《文物保护与考古科学》
北大核心
2021年第1期17-25,共9页
Sciences of Conservation and Archaeology
关键词
高光谱成像
彩绘铜车马
纹样数字复原
颜料识别
Hyperspectral imaging
Painted bronze chariot
Pattern digital restoration
Pigment identification