Manfredi Marcello, Bearman Greg, Williamson Greg, Kronkright Dale, Doehne Eric, Jacobs Megan, Marengo Emilio
Department of Sciences and Technological Innovation, University of Easter Piedmont, Viale T. Michel 11, Alessandria 15121, Italy.
ANE Image, Pasadena, CA 91104, USA.
Sensors (Basel). 2014 Jul 9;14(7):12271-84. doi: 10.3390/s140712271.
In this paper we propose a reliable surface imaging method for the non-invasive detection of morphological changes in paintings. Usually, the evaluation and quantification of changes and defects results mostly from an optical and subjective assessment, through the comparison of the previous and subsequent state of conservation and by means of condition reports. Using quantitative Reflectance Transformation Imaging (RTI) we obtain detailed information on the geometry and morphology of the painting surface with a fast, precise and non-invasive method. Accurate and quantitative measurements of deterioration were acquired after the painting experienced artificial damage. Morphological changes were documented using normal vector images while the intensity map succeeded in highlighting, quantifying and describing the physical changes. We estimate that the technique can detect a morphological damage slightly smaller than 0.3 mm, which would be difficult to detect with the eye, considering the painting size. This non-invasive tool could be very useful, for example, to examine paintings and artwork before they travel on loan or during a restoration. The method lends itself to automated analysis of large images and datasets. Quantitative RTI thus eases the transition of extending human vision into the realm of measuring change over time.
在本文中，我们提出了一种可靠的表面成像方法，用于非侵入性检测绘画作品中的形态变化。通常，通过比较绘画作品先前和后续的保存状态并借助状况报告，对变化和缺陷的评估及量化大多基于光学和主观评估。利用定量反射变换成像（RTI），我们通过一种快速、精确且非侵入性的方法获得了绘画表面几何形状和形态的详细信息。在绘画作品遭受人工损伤后，获取了关于其劣化的准确且定量的测量数据。使用法向量图像记录形态变化，而强度图成功地突出、量化并描述了物理变化。我们估计，考虑到绘画作品的尺寸，该技术能够检测到略小于0.3毫米的形态损伤，而这种损伤肉眼很难察觉。例如，这种非侵入性工具对于在绘画作品外借或修复期间对其进行检查可能非常有用。该方法适用于对大图像和数据集进行自动分析。因此，定量RTI有助于将人类视觉扩展到测量随时间变化的领域。
