Proc. SPIE. 7506, 2009 International Conference on Optical Instruments and Technology: Optical Systems and Modern Optoelectronic Instruments
KEYWORDS: Near infrared, Principal component analysis, Statistical analysis, Diffuse reflectance spectroscopy, Data modeling, Pattern recognition, Spectroscopy, Manufacturing, Near infrared spectroscopy, Analytical research
Near-infrared (NIR) diffuse reflectance spectroscopy and pattern recognition techniques are applied to develop a fast
identification method of Jinhua ham. The samples are collected from different manufactures and they are nineteen Jinhua
ham samples and four Xuanwei ham samples. NIR spectra are pretreated with second derivative calculation and vector
normalization. The pattern recognition techniques which are cluster analysis, conformity test and principal component
analysis (PCA) are separately used to qualify Jinhua ham. The three methods can all distinguish Jinhua ham successfully.
The result indicated that a 100 % recognition ration is achieved by the methods and the PCA method is the best one.
Overall, NIR reflectance spectroscopy using pattern recognition is shown to have significant potential as a rapid and
accurate method for identification of ham.
Tunable diode laser absorption spectroscopy (TDLAS) is a method to detect trace-gas qualitatively or quantitatively
based on the tunable characteristic of the diode laser to obtain the absorption spectroscopy in the characteristic
absorption region. The concentration of CO is measured by tunable diode laser absorption spectroscopy (TDLAS)
technology in this paper. The experimental results of measurement signals are inversely processed by applying the
overall second harmonic least squares data processing algorithm. The experimental results indicate that the signal
strength of the second harmonic spectrum changes with CO concentration. But the widths of the 2f lineshapes have not
changed. The components that influence the uncertainty of measurement results during measuring CO concentration by
TDLAS are analyzed and the mathematic model is built. The standard uncertainty of components and evaluation of
uncertainty of measurement results are given with the direct evaluation method in detail. The evaluation results indicate
that the major factors affecting measurement uncertainty are the indicating value uncertainty of the apparatus,
concentration definite value uncertainty of calibrating gas.