Premium
Discrimination of E uterpe oleracea M art. (Açaí) and E uterpe edulis M art. (Juçara) Intact Fruit Using Near‐Infrared ( NIR ) Spectroscopy and Linear Discriminant Analysis
Author(s) -
Dall' Acqua Yara Gurgel,
Cunha Júnior Luis Carlos,
Nardini Viviani,
Lopes Valquira Garcia,
Pessoa José Dalton da Cruz,
Almeida Teixeira Gustavo Henrique
Publication year - 2015
Publication title -
journal of food processing and preservation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.511
H-Index - 48
eISSN - 1745-4549
pISSN - 0145-8892
DOI - 10.1111/jfpp.12536
Subject(s) - linear discriminant analysis , food science , food industry , chemistry , botany , horticulture , biology , mathematics , statistics
Abstract As the verification of authenticity of agricultural foods has become a potential application of spectroscopic methods, this study aimed to use near‐infrared ( NIR ) spectroscopy associated with linear discriminant analysis ( LDA ) to discriminate intact fruit of the species E uterpe oleracea M art. (açaí) and E uterpe edulis M art. (juçara). One hundred sixty‐eight açaí fruits from 17 genotypes and 200 fruits from 20 juçara genotypes were investigated and two groups were used for discriminating the species: calibration (294 fruits) and prediction (74 fruits). L *, a *, b *‐ PCA , and NIR‐PCA of spectra pretreated with multiplicative scatter correction ( MSC ) showed the best results. The L *, a *, b *‐ LDA model resulted in 96.3% correct classification and 93.2% prediction accuracy of the external validation group. NIR spectra pretreated with MSC had 98% correct classification and 97.3% prediction accuracy. NIR spectroscopy associated with LDA is a reliable method for the discrimination of intact açaí and juçara fruit. Practical Applications Recently, there has been an increasing interest by food industry and producers to assure consumers the authenticity of their products. Many food properties are related to individual compounds and their active chemical ingredients, such as essential oils, terpenoids, flavonoids, phenolic compounds, amino acids and organic acids. The Euterpe genus has many species of economic interest and E. oleracea (açaí or assai), E. edulis (juçara) and E. precatoria are among the most important species in the agribusiness sector. As the fruit of these species are morphologically similar, it is important to develop instrumental methods to sort these fruit at the convey belt in order to ensure authenticity and near‐infrared (NIR) spectroscopy can be used for that. Based on our results, the NIR spectroscopy and chemometrics provide a useful approach for authenticating fruit from Euterpe genus and can be used by the food industry to identify adulteration.