University of California

For Abstracts

Use of Digital Image Processing for Evaluation of Translucency in Fresh Cut ‘PÉROLA’ Pineapple Coated with Biofilms

Renato Pereira Lima (CCA/UFPB, BRAZIL), Silvanda de Melo Silva* (PPGA/CCA/UFPB, BRAZIL), Renato Lima Dantas (UACTA/UFCG), Ana Lima Dantas (PPGA/CCA/UFPB), Alex Sandro Bezerra de Sousa (CCA/UFPB), Walter Esfrain Pereira (PPGA/CCA/UFPB), Rejane Maria Nunes Mendonça (PPGA/CCA/UFPB)

The translucency in 'Perola' pineapple, characterized by water-soaking areas in the flesh, intensifies during storage of fresh cut product causing loss of quality. The evaluation of the pineapple translucency has been performed by subjective scales, grading the evolution of the translucent area of the cross section of fruit. However, the digital image processing, using free software like ImageJ®, can allow the extraction of more precise information on the translucency considering, for instance, the tone of image coloration, the areas in which it occurs, and the pixels in the RGB channels. Therefore, this study proposed a method for evaluation of translucency in fresh cut ‘Pérola´ pineapple using the digital image processing (DIP). 'Perola' pineapple was cut in 10 mm-thick slices, which were coated with cassava starch (3%), cassava starch (3%) + fennel oil (0.025%), cassava starch (2%) + sodium alginate (1%) + fennel oil (0.035%), cassava starch (3%) + glycerol (1%) + ascorbic acid (0.5%), and stored for 12 days at 5 degrees C and 90% RH. The DIP was developed using variations of hue angle, saturation, and brightness by the ImageJ® software. The correlation of the subjective evaluation with the DIP was highly significant (r = 0.90 **), i.e., it was efficient at determining the translucency in ‘Perola’ pineapple slices. A predictor model was developed using partial least squares analysis. This model, which included the coloration of the slices and the RGB values, had a high linear coefficient of determination (r = 0.96 significance).

Top of page

Webmaster Email: