Lavine Barry K, Fasasi Ayuba, Mirjankar Nikhil, Sandercock Mark
Department of Chemistry, Oklahoma State University, Stillwater, OK 74078, USA.
Department of Chemistry, Oklahoma State University, Stillwater, OK 74078, USA.
Talanta. 2014 Feb;119:331-40. doi: 10.1016/j.talanta.2013.10.066. Epub 2013 Nov 12.
Search prefilters developed from spectral data collected on two 6700 Thermo-Nicolet FTIR spectrometers were able to identify the respective manufacturing plant and the production line of an automotive vehicle from its clear coat paint smear using IR transmission spectra collected on a Bio-Rad 40A or Bio-Rad 60 FTIR spectrometer. All four spectrometers were equipped with DTGS detectors. An approach based on instrumental line functions was used to transfer the classification model between the Thermo-Nicolet and Bio-Rad instruments. In this study, 209 IR spectra of clear coat paint smears comprising the training set were collected using two Thermo-Nicolet 6700 IR spectrometers, whereas the validation set consisted of 242 IR spectra of clear coats obtained using two Bio-Rad FTIR instruments.
从两台赛默飞世尔尼高力6700傅里叶变换红外光谱仪收集的光谱数据开发的搜索预过滤器，能够使用在伯乐40A或伯乐60傅里叶变换红外光谱仪上收集的红外透射光谱，从汽车清漆涂片识别出该汽车的制造工厂和生产线。所有四台光谱仪均配备了DTGS探测器。采用一种基于仪器谱线函数的方法在赛默飞世尔尼高力和伯乐仪器之间转移分类模型。在本研究中，使用两台赛默飞世尔尼高力6700红外光谱仪收集了209个构成训练集的清漆涂片红外光谱，而验证集由使用两台伯乐傅里叶变换红外仪器获得的242个清漆红外光谱组成。
