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

Abstract

This master's project will investigate the application of near-infrared spectral images (NIR-HSI) for differentiation and determination of the composition of different varieties of cocoa beans. Chemical composition and physical-chemical parameters such as color and pH will be evaluated for characterization of the samples. The images will be acquired and the region of interest to be analyzed will be isolated to obtain the spectra. The most relevant wavelengths in spectral measurements will be identified through variable selection methods to be used in predictive models, in order to reduce the volume of data for processing, enabling applications in the food industry. This work will be developed within a Young Researcher Project, whose proposal is to develop innovative methods of food evaluation with potential for medium-term application in industry. (AU)