Prediction of colour of European Emmental cheeses by using near infrared spectroscopy: a feasibility study

• Published in: European food research & technology 2007-11, Vol.226 (1-2), p.63-69
• Main Authors: Pillonel, Laurent, Dufour, Eric, Schaller, Emmanuelle, Bosset, Jacques-Olivier, De Baerdemaeker, Josse, Karoui, Romdhane
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; Chemometric; Color; Correlation coefficient; Dairy products; Emmental cheese; Feasibility studies; Food industries; Fundamental and applied biological sciences. Psychology; Infrared spectroscopy; Mean square errors; Milk and cheese industries. Ice creams; Near infrared; Principal components analysis; Spectrum analysis
• ISSN: 1438-2377; 1438-2385
• DOI: 10.1007/s00217-006-0509-4

Near infrared (NIR) spectroscopy was used to predict colour of European Emmental cheese samples. Colour values (L, a and b) were measured on 20 Emmental cheese samples using a Hunter-lab D25-D-2 optical head in the system according to Hunter to determine L (brightness), a (green-red component) and b (blue-yellow component). The diffuse reflectance of the investigated cheeses was also determined by a Büchi NIR Lab N-200 spectrometer using a rotating measuring cell in the range of 1000-2500 nm. The best results for L-value (squared correlation coefficient (R ²) = 0.56, root mean square error of cross-validation (RMSECV) = 0.76, ratio of prediction deviation (RPD) = 1.89 and range error ratio (RER) = 7.91), a-value (R ² = 0.72, RMSECV = 0.15, RPD = 1.98 and RER = 7.6) and b-value (R ² = 0.82, RMSECV = 0.52, RPD = 2.56 and RER = 9.42) were obtained when the first 12 principal components (PCs) of the principal component analysis (PCA) applied on normalised NIR spectra were used. It can be concluded that NIR spectroscopy could be used to predict b-value. The a- and L-values can also be predicted from NIR technique with approximate quantitative prediction.