高光谱成像技术在充胶处理宝石鉴定中的应用——以翡翠和绿松石为例

Application of Hyperspectral Imaging Technique in Identification of Polymer-Impregnated Gemstone:Taking Jadeite and Turquoise as Example

20世纪80年代,一种传统影像学和光谱学结合形成的新型技术——高光谱成像技术开始兴起并迅速发展,使用高光谱成像测试仪可以得到集图像数据和光谱数据于一体的三维数据信息,而该项技术在宝石学方面还未得到广泛应用。目前主要依赖红外光谱对经充胶处理的宝石进行无损检测,其测试结果会受到样品表面抛光程度及样品透明度的影响,同时,对胶物质分布及充胶程度较难得知。高光谱成像技术对样品抛光程度及透明度要求不高,在一定程度上能避免红外光谱由于抛光程度、透明度对测试结果的影响,且能得知充胶物质分布及充胶程度。采用小型高光成像仪对充胶翡翠、有机充胶绿松石和天然翡翠、绿松石进行测试,并将结果用ENVI软件进行处理。结果显示,1 000～2 500 nm(近红外波段)翡翠和绿松石样品的成像光谱信息,其中翡翠充胶处理样品出现1 733、2 208、2 313 nm处的吸收,绿松石充胶样品在2 000～2 300 nm范围多吸收峰特征可与天然绿松石区别,推测其中1 733 nm处的吸收为有机胶所致。高光谱成像光谱测试结果经ENVI软件处理,得到充胶分布和充胶程度图。

In the 1980s, hyperspectral imaging technique was developed rapidly, which was formed by the combination of traditional imaging technique and spectroscopy. Using hyperspectral imaging tester, 3D data information can be obtained by integrating image data and spectral data. But this technique has not been widely used in gemmology. At present, it mainly depends on FTIR to achieve non-destructive identification of polymer-impregnated gemstones, and sometimes the results are affected by the surface polishing and the transparency of the sample. FTIR is difficult to find out the distribution and degree of the filling. However, the hyperspectral imaging techique can cover those shortages of FTIR. In the papers, hyperspectral imaging technique was used to compare polymer-impregnated and natural jadeite and turquoise samples, and ENVI software was used to process the results. The results are as follows: The imaging spectral information of jadeite and turquoise samples in 1 000-2 500 nm (in NIR band) was obtained. The imaging spectral information of jadeite and turquoise samples in 1 000-2 500 nm band was obtained. The absorption of polymer-impregnated jadeite at 1 733、2 208、2 313 nm was observed. The diagnostic of absorption in 2 000- 2 300 nm of polymer-impregnated turquoise could be distinguished from those of natural turquoise, which is inferred that the absorption of 1 733 nm is caused by polymer. By ENVI processing, the distribution of polymer and the degree diagram were obtained and shown distinct.