Multivariate statistical analyses applied to NIR spectroscopy and digital image analyses for food products

Abstract

The current work plan proposes the application of chemometric methods and multivariate analysis applied to near infrared (NIR) spectral information and digital images (RGB) for classification and authentication of cocoa butter. The samples will be characterized for composition and physical-chemical attributes such as color and pH. Digital images and spectral information will be obtained using portable equipment for further analysis. The most relevant wavelengths in spectral measurements will be identified for use in simplified predictive models. This step allows the development of other techniques, such as spectral imaging, as potential applications in the form of process analytical technologies (PAT) in the food industry. Similarly, parameters obtained from the digital images will be used as predictors for the identification of samples. This work will be developed within a Young Researcher Project, which proposal is to develop innovative methods as process analytical technologies in the food industry. (AU)