Authenticity Identification of Mutton Based on Electronic Nose and Visible/Near-infrared Spectroscopy
Article
Figures
Metrics
Preview PDF
Reference
Related
Cited by
Materials
Abstract:
In order to identify the minced mutton products adulterated with duck rapidly and accurately, an electronic nose combined with visible/near-infrared spectroscopy technology was used to realize effective identification. A total of 174 samples of minced mutton adulterated with duck in different proportions were prepared, and electronic nose data and reflection spectra in the wavelength ranges of 200~1 100 nm and 900~1 700 nm were acquired, respectively. Two-classification qualitative discrimination and six-classification quantitative detection methods were used to establish the support vector machine (SVM) and partial least squares (PLS) qualitative and quantitative discrimination models, respectively. Subsequently, the six-classification optimal models were used for prediction based on two spectral bands. The electronic nose detected and identified the six groups through the odor difference. The contents of volatile aroma components such as terpenes, aromatic compounds, and organic sulfides in mutton were higher than those in duck. The PLS models based on two-classification methods and spectral data of two bands were superior to the SVM models, and the total discriminant accuracy was more than 96%. The best spectral pretreatment method was multiplicative scatter correction, and the final optimal model predicted well. In summary, an electronic nose combined with visible/near infrared spectroscopy can effectively identify the mutton samples adulterated with duck, thus providing technical support for rapid and nondestructive identification of mutton adulteration.
