Study on the Characteristic Spectral Properties for Species Identification of Animal-Derived Feedstuff Using Fourier Transform Infrared Spectroscopy

Abstract

The objective of the present study was to explore the effective spectral bands related to lipid characteristics in spectra of raw animal-derived feedstuff and figure out which marked spectral regions (single or combined) contributed more to species discrimination. A total of 82 meat and bone meals, including porcine, poultry, bovine, ovine, and fish, were studied. Raw materials, extracted lipid, and defatted samples were simultaneously analyzed and calculated using Fourier transform infrared (FT-IR) spectroscopy in combination with chemometric methods. Taking the spectra of lipid as references, five marked spectral regions considered the main lipid characteristic regions were found in the raw animal-derived feedstuff spectra. In the study, single and combined marked spectral bands were investigated and proved to have better performance than the whole spectra of raw terrestrial animal-derived feedstuff and fishmeal. For the discrimination of five animal species, the regions of 1800–1650 cm^–1, 1500–1330 cm^–1, 1260–1060 cm^–1, and 790–640 cm^–1 presented better results; for the classification of three categories, the regions of 3100–2800 cm^–1, 1800–1650 cm^–1, and 1500–1330 cm^–1 showed the best results.