Application of Various Pre-Processing Techniques on Infrared (IR) Spectroscopy Data for Classification of Different Ghee Samples

When dealing with IR spectroscopy, the preprocessing of spectral data is considered as one of the most important parts of chemometrics modeling. Due to any uncontrollable physical variations may lead to an additive, multiplicative and wavelength-dependent scattering effects in the recorded spectra. Pre-processing techniques basically are required to remove these scattering effects from the spectra and subsequently improve the further quantitative and qualitative analysis. Most popular pre-processing techniques are; baseline correction, smoothing of the spectra, normalization, scattering correction and spectral derivatives. This paper begins with the theoretical and mathematical foundation of various pre-processing techniques used for IR spectroscopy. Then a qualitative analysis is performed by applying these techniques to the spectral data collected using various samples of ghee. The comparison of various preprocessing is obtained by modeling of the data using Principle Component Analysis (PCA) and then the k-means clustering algorithm.