关键词: MA-XRF clustering dimensionality reduction elemental maps painting stratigraphy panel painting pigments

来  源:   DOI:10.3390/jimaging8050147

Abstract:
X-ray fluorescence (XRF) spectrometry has proven to be a core, non-destructive, analytical technique in cultural heritage studies mainly because of its non-invasive character and ability to rapidly reveal the elemental composition of the analyzed artifacts. Being able to penetrate deeper into matter than the visible light, X-rays allow further analysis that may eventually lead to the extraction of information that pertains to the substrate(s) of an artifact. The recently developed scanning macroscopic X-ray fluorescence method (MA-XRF) allows for the extraction of elemental distribution images. The present work aimed at comparing two different analysis methods for interpreting the large number of XRF spectra collected in the framework of MA-XRF analysis. The measured spectra were analyzed in two ways: a merely spectroscopic approach and an exploratory data analysis approach. The potentialities of the applied methods are showcased on a notable 18th-century Greek religious panel painting. The spectroscopic approach separately analyses each one of the measured spectra and leads to the construction of single-element spatial distribution images (element maps). The statistical data analysis approach leads to the grouping of all spectra into distinct clusters with common features, while afterward dimensionality reduction algorithms help reduce thousands of channels of XRF spectra in an easily perceived dataset of two-dimensional images. The two analytical approaches allow extracting detailed information about the pigments used and paint layer stratigraphy (i.e., painting technique) as well as restoration interventions/state of preservation.
摘要:
X射线荧光(XRF)光谱法已被证明是一个核心,非破坏性的,文化遗产研究中的分析技术主要是因为它具有非侵入性和快速揭示被分析文物的元素组成的能力。能够比可见光更深地渗透到物质中,X射线允许进一步分析,这最终可以导致提取与伪影的衬底有关的信息。最近开发的扫描宏观X射线荧光法(MA-XRF)允许提取元素分布图像。本工作旨在比较两种不同的分析方法,以解释在MA-XRF分析框架中收集的大量XRF光谱。以两种方式分析测量的光谱:仅光谱方法和探索性数据分析方法。应用方法的潜力展示在著名的18世纪希腊宗教面板绘画上。光谱方法分别分析每个测量的光谱,并导致构建单元素空间分布图像(元素图)。统计数据分析方法导致将所有光谱分组为具有共同特征的不同簇,而随后的降维算法有助于在易于感知的二维图像数据集中减少数千个通道的XRF光谱。这两种分析方法允许提取有关所用颜料和油漆层地层学的详细信息(即,绘画技术)以及恢复干预/保存状态。
公众号