Premium
Potential use of multispectral imaging technology to identify moisture content and water‐holding capacity in cooked pork sausages
Author(s) -
Ma Fei,
Zhang Bin,
Wang Wu,
Li Peijun,
Niu Xiangli,
Chen Conggui,
Zheng Lei
Publication year - 2017
Publication title -
journal of the science of food and agriculture
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.782
H-Index - 142
eISSN - 1097-0010
pISSN - 0022-5142
DOI - 10.1002/jsfa.8659
Subject(s) - partial least squares regression , multispectral image , water holding capacity , water content , visualization , moisture , multivariate statistics , chemistry , environmental science , biological system , food science , computer science , artificial intelligence , machine learning , geotechnical engineering , organic chemistry , engineering , biology
BACKGROUND The traditional detection methods for moisture content (MC) and water‐holding capacity (WHC) in cooked pork sausages (CPS) are destructive, time consuming, require skilled personnel and are not suitable for online industry applications. The goal of this work was to explore the potential of multispectral imaging (MSI) in combination with multivariate analysis for the identification of MC and WHC in CPS. RESULTS Spectra and textures of 156 CPS treated by six salt concentrations (0–2.5%) were analyzed using different calibration models to find the most optimal results of predicting MC and WHC in CPS. By using the fused data of spectra and textures, partial least squares regression models performed well for determining the MC and WHC, with a correlation coefficient ( r ) of 0.949 and 0.832, respectively. Additionally, their spatial distribution in CPS could be visualized via applying prediction equations to transfer each pixel in the image. CONCLUSION Results of satisfactory detection and visualization of the MC and WHC showed that MSI has the potential to serve as a rapid and non‐destructive method for use in sausage industry. © 2017 Society of Chemical Industry