Browsing Food Quality & Sensory Science by Subject "NIR"
Now showing items 1-2 of 2
Application of class-modelling techniques to near infrared data for food authentication purposesFollowing the introduction of legal identifiers of geographic origin within Europe, methods for confirming any such claims are required. Spectroscopic techniques provide a method for rapid and non-destructive data collection and a variety of chemometric approaches have been deployed for their interrogation. In this present study, class-modelling techniques (SIMCA, UNEQ and POTFUN) have been deployed after data compression by principal component analysis for the development of class-models for a set of olive oil and honey. The number of principal components, the confidence level and spectral pre-treatments (1st and 2nd derivative, standard normal variate) were varied, and a strategy for variable selection was tried. Models were evaluated on a separate validation sample set. The outcomes are reported and criteria for selection of the most appropriate models for any given application are discussed.
Near Infrared Spectroscopy in the Food Industry: A Tool of Quality Management.Near infrared (NIR) spectroscopy is a rapid, non-destructive analytical technique which has been used in the food and agriculture industries for almost 20 years. Ireland was one of the first countries in the world to adopt this method for national trading purposes and the grain trade has used it for off-farm and in-process analysis since 1981. However, other sectors have been slower to realise its potential and as part of a process of demonstrating the role which it may play in monitoring quality in a range of food industry applications, a programme of research and development has been on-going within Teagasc and its predecessor An Foras Talúntais. NIR spectroscopy provides the food processor with information. This information may describe how much of a given substance is present in a mixture or how the overall quality of the substance compares to a reference material e.g. a previous batch of raw material, finished goods or a competitor’s product. This report provides some examples of precompetitive R&D on representative qualitative and quantitative problems in a range of foods and food ingredients. The use of NIR spectra collected within 24 hours of slaughter to predict beef tenderness 14 days later shows considerable promise. Non-destructive monitoring of flesh composition in farmed salmon has paved the way for the efficient use of expensive feed materials while the content of each species in binary mixtures of minced beef and lamb has been accurate enough to suggest the use of NIR spectroscopy as a rapid screening tool by regulatory agencies, food processors and retailers. Classification of a range of food ingredients (including skim milk powder and flour) into one of a number of functionally-discrete categories has been successfully achieved with levels of accuracy high enough to warrant immediate industry utilisation i.e. greater than 90% for skim milk powders and 97% in the case of flour. Species confirmation in a number of raw minced meats (chicken, turkey, pork, beef and lamb) has been achieved with over 90% accuracy in feasibility studies. Calibrations transferred from one NIR instrument to another lose accuracy because of differences in instrument construction, sample presentation and other factors. A research effort has recently been applied to this problem of transferability and results are available for both scanning and fixed filter instruments. The success achieved opens the way for using NIR results obtained in different companies or countries as an uncontested basis for trade.