Prediction of dry matter content of recently harvested ‘Hass’ avocado fruits using hyperspectral imaging

Abstract

BACKGROUND‘Hass’ avocado consumption is increasing due to its organoleptic properties, so it is necessary to develop new technologies to guarantee export quality. Avocado fruits do not ripen on the tree, and the visual classification of its maturity is not accurate. The most commonly used fruit maturity indicator is the percentage of dry matter (DM). The aim of this research was to investigate a non‐destructive method with hyperspectral images to predict the percentage of DM of fruits across the spectral range of 400–1000 nm.RESULTSNo correlation between fruit weight and color with the percentage of DM was found in the study area. Cross‐validation efficiency of different data sources, including the spectrum extraction zone (the center, a line from the peduncle to the base, and the whole fruit) and the average of one or two fruit faces, was compared. Four linear regression models were compared. Data of the whole fruit and average of both sides per fruit using a support vector machine regression were selected for the prediction test. Following the cross‐validation concept, five sets of calibration and test data were selected and optimized for calibration. The best test prediction set comprised an R2 = 0.9, a root‐mean‐square error of 2.6 g kg−1 DM, a Pearson correlation of 0.95, and a ratio of prediction to deviation of 3.2.CONCLUSIONSThe results of the study indicate that hyperspectral images allow classifying export fruits and making harvesting decisions. © 2020 Society of Chemical Industry

MEDUSA: An open-source and webcam based multispectral imaging system

“…Some of these differences might be explained by differences in fruit ripening. Previous research has also shown chances for non-contact estimation of avocado ripeness at harvest, as reported by [23] , which made a comprehensive study on avocado ripening detection by using a scientific grade hyperspectral camera, industrial lighting and refined image analysis algorithms in an external software platform.…”

Postharvest Treatments of Hass Avocado (Persea americana Mill.) and Estimation of Its Quality Using Hyperspectral Imaging (HSI)

“…Hyperspectral images of the avocados were acquired in the visible/near-infrared spectrum (VIS/NIR, 400−1000 nm) with a 12-bit hyperspectral camera (Pika XC2, Resonon Inc., Bozeman, MT) using the lighting system described by Vega, Sandoval, and Reina. 8 The camera was calibrated following the manufacturer's guidelines. 31 Data acquisition was performed using Spectronon software (v2.118, Resonon Inc.).…”

“…11 Coupling multidimensional spectral analysis with machine learning techniques allows exploiting and de-entangling dense spectral data to rapidly estimate outcomes (e.g., changes in relevant quality attributes and extent of spoilage) for alternative scenarios. 12 This approach has proven effective in estimating avocados' physiological maturity 8 and some postharvest quality attributes. 13 However, additional efforts are required to develop monitoring techniques that include consumer perception within their development.…”

“…These sampling limitations often result in an incomplete representation of a lot due to the high heterogeneity between fruits . To overcome these problems, nondestructive methods, such as hyperspectral imaging, have been suggested as a better and more complete monitoring approach. − Hyperspectral images are composed of three dimensions, two spatial and one spectral, that usually have ten to hundreds of continuous bands. The three-dimensional hyperspectral cubes, also called hypercubes (x, y, λ), are mainly obtained by scanning points, lines, or areas .…”

Avocado fruit sorting by hyperspectral images

“…В работе [13] показана возможность классификации плодов авокадо по степени зрелости использованием HSI в диапазоне 300-900 нм. Разработка калибровочных моделей для определения физико-химических свойств плодов авокадо продемонстрирована в работе [14].…”