Description of the subject. Phenolic compounds (PCs) are the most abundant secondary metabolites in plants. This work was part of a study that sought to develop rapid screening FT-Raman methods for identifying and quantifying classes and/or types of PCs in the dry extracts of plant products. Objectives. Fourier transform Raman spectroscopy (FT-Raman) coupled with chemometric tools was used to characterize and discriminate four families of phenolic compounds: two important classes of phenolic acids, hydroxybenzoic and hydroxycinnamic acids, as well as their derivatives and flavonoids. Method. For this study, 25 standards of phenolic compounds were used (47 standards in total, taking account of the different brands). Repeatability and reproducibility studies were conducted to verify the Raman assignments of gallic acid. Raman characterization with the most significant spectral bands of phenolic compounds was done using spectra ranging from 1,800 to 50 cm-1. A Fisher test was applied to pre-processed Raman spectra (SNV) and 20 Raman scattering signals were used to differentiate each class of phenolic compounds. Results. Hydroxybenzoic acids, hydroxycinnamic acids, their derivatives and flavonoids have been characterized by Raman spectroscopy; bands have been identified and differentiated within and between groups. Conclusions. The scattering intensities located around 1,600-1,699, 1,300-1,400 and below 200 cm-1 were responsible for differentiating 100% of phenolic compound families, classes and subclasses.

Cocoa hybridisation generates new varieties which are resistant to several plant diseases, but has individual chemical characteristics that affect chocolate production. Image analysis is a useful method for visual discrimination of cocoa beans, while deep learning (DL) has emerged as thede factotechnique for image processing . However, these algorithms require a large amount of data and careful tuning of hyperparameters. Since it is necessary to acquire a large number of images to encompass the wide range of agricultural products, in this paper, we compare a Deep Computer Vision System (DCVS) and a traditional Computer Vision System (CVS) to classify cocoa beans into different varieties. For DCVS, we used a Resnet18 and Resnet50 as backbone, while for CVS, we experimented traditional machine learning algorithms, Support Vector Machine (SVM), and Random Forest (RF). All the algorithms were selected since they provide good classification performance and their potential application for food classification A dataset with 1,239 samples was used to evaluate both systems. The best accuracy was 96.82% for DCVS (ResNet 18), compared to 85.71% obtained by the CVS using SVM. The essential handcrafted features were reported and discussed regarding their influence on cocoa bean classification. Class Activation Maps was applied to DCVS’s predictions, providing a meaningful visualisation of the most important regions of the images in the model.

Determination of nutritionally important trans MUFA, CLA, and OBCFA milk fatty acids (often present in amounts lower than 1.0 g/100 g of total fat) using fast and nondestructive analytical methods would enhance their use as diagnostic tools in dairy herd and human health management. Here, PLS regression using ATR/FTIR spectra indicated potential for determination of trans-11 C18:1 and trans-12 C18:1 (R_cv² ≥ 0.80), and trans-9 C18:1 in very minor concentration (R_cv² > 0.82), as well as anteiso C15:0 (R_cv² = 0.57) and iso C17:0 (R_cv² = 0.61). Furthermore, the main cis-9,trans-11 CLA isomer was predicted well despite the high trans MUFA concentration. Differentiation between the CLA and the trans MUFA signals was evident (based on specific cis/trans bands), and branched-chain saturated fatty acid methyl esters revealed specific iso and anteiso ATR/FTIR absorbance bands. None of the minor FA PLS results with FT-NIR showed interesting potential, except satisfactory predictions for trans-9 C18:1 and cis-9,trans-11 CLA. Overall, ATR/FTIR resulted in better calibrations and provided more specific information for determination of minor milk fatty acids.