Determination of the moisture content of fresh meat using visible and near-infrared spatially resolved reflectance spectroscopy. (October 2017)

• Record Type: Journal Article
• Authors: Wang, Jian-Xu; Fan, Li-Feng; Wang, Hai-Hua; Zhao, Peng-Fei; Li, Hao; Wang, Zhong-Yi; Huang, Lan
• Is Part Of: Biosystems engineering. Volume 162(2017)
• Page Start: 40
• Page End: 56
• Publication Date: 2017-10
• Subjects: Spatially resolved reflectance spectroscopy -- Intact pork -- Moisture -- Partial-least-squares regression -- Successive-projection algorithm -- Variable selection; Bioengineering -- Periodicals; Agricultural engineering -- Periodicals; Biological systems -- Periodicals; Génie rural -- Périodiques; Systèmes biologiques -- Périodiques; 631
• DOI: 10.1016/j.biosystemseng.2017.07.004
• Languages: English
• ISSNs: 1537-5110

Abstract

Abstract : The moisture content of intact pork meat was determined employing near-infrared non-destructive reflectance spectroscopy. The effects of various factors, namely the shape of the probe, the contact between the probe and sample and the modelling method, were also evaluated. Wavelengths in the range from 600 to 1100 nm, two spectra collection methods and three corresponding data processing methods were considered. Spectra were collected using a conventional Y-shaped optical fibre probe and an optical probe for steady-state spatially resolved detection. In the experiments, we can obtain the spectral information from the depth approximately 3 mm–4 mm or 4 mm–5 mm when the separation of the detector and the source is 6 mm or 9 mm, respectively. Models were established using data obtained with the steady-state spatially resolved probe (41 samples) and the Y-shaped optical probe (28 samples). Using the three-wavelength and Partial-least-squares regression (PLSR) calibration methods, the Y-shaped probe provided models with R 2 values of 0.0097 and 0.36 respectively, whereas the steady-state spatially resolved probe provided models with R 2 values of 0.3692 and 0.7769 respectively. Applying a successive-projection algorithm combined with multiple linear regressions to the data of the steady-state spatially resolved probe, the R 2 of the obtained model was 0.8078. Near-infrared steady-state spatially resolved spectroscopy is thus an effective technique for predicting the moisture content of intact pork. Highlights: Non-destructive moisture measurement using spatially-resolved NIR spectroscopy. Estimating detection depth of the intact pork by using Monte Carlo simulation. Designing a steady spatially-resolved detection probe. Three algorithms were used for predicting moisture content of intact pork meat. Selecting effective wavelengths with SPA-MLR.