SEGMENTATION AND EDGE DETECTION ALGORITHMS FOR DIGITAL IMAGE ANALYSIS IN BERRIES SUBJECTED TO MICROWAVE DRYING

VASCO, MARÍA FLORENCIA; LAURA CAMPAÑONE; GAMBOA, JULIANA

Congreso; 34TH EUROPEAN FEDERATION FOR FOOD SCIENCE AND TECHNOLOGY (EFFOST).; 2020

ELSEVIER

The digital image analysis algorithms presented in this work showed a great potential to obtain valuable information about area and color reductions during a microwave-assisted drying process, highlighting the efficiency of this technique as food quality in-line monitoring tool. The Otsu?s algorithm with a previous Gaussian filtering and the HSV thresholding methods showed good segmentation results for the MW plate. However, at long drying times, over-segmented areas were obtained which suggested the need of an effective prior edge detection method for monitoring area reduction. In future work, the improvement of the Canny?s edge detection algorithm with morphological operations and the study of other algorithms could be suitable strategies for a more accurate determination of the real surface area during drying processing.