Application of Imaging and Artificial Intelligence for Quality Monitoring of Stored Black Currant (Ribes nigrum L.)

Abstract

The objective of this study was to assess the influence of storage under different storage conditions on black currant quality in a non-destructive and inexpensive manner using image processing and artificial intelligence. Black currants were stored at a room temperature of 20 ± 1 °C and a temperature of 3 °C (refrigerator). The images of black currants directly after harvest and fruit stored for one and two weeks were obtained using a digital camera. Then, texture parameters were computed from the images converted to color channels R (red), G (green), B (blue), L (lightness component from black to white), a (green for negative and red for positive values), b (blue for negative and yellow for positive values), X (component with color information), Y (lightness), and Z (component with color information). Models for the classification of black currants were built using various machine learning algorithms based on selected textures for RGB, Lab, and XYZ color spaces. Models built using the IBk, multilayer perceptron, and multiclass classifier for textures from RGB color space, and the IBk algorithm for textures from Lab color space distinguished unstored black currants and samples stored in the room for one and two weeks with an average accuracy of 100%, and the kappa statistic and weighted averages of precision, recall, Matthews correlation coefficient (MCC), receiver operating characteristic (ROC) area, and precision–recall (PRC) area equal to 1.000. This indicated a very distinct change in the external structure of the fruit after the first week and more and more visible changes in quality with increasing storage time. A classification accuracy reaching 98.67% (multilayer perceptron, Lab color space) for the samples stored in the refrigerator may indicate smaller quality changes caused by storage at a low temperature. The approach combining image textures and artificial intelligence turned out to be promising to monitor the quality changes in black currants during storage.

Enhancing Pigment Phenotyping and Classification in Lettuce through the Integration of Reflectance Spectroscopy and AI Algorithms

“…However, it has been noted that these methods have limitations, such as low accuracy and biases in the output. On average, our findings indicate that 11.5% of the STEPW and VIP values were shared among the range of data, with VIP values ranging from 5.1 to 20.6%, as reported in [ 1 , 26 , 31 ].…”

“…In lettuce plants, they can be used to analyse hyperspectral data or ATR-FTIR spectroscopy data to accurately classify lettuce varieties, and to predict crop yield or quality [ 15 , 23 ]. Deep learning is a specific type of machine learning that can be particularly useful for these tasks, as it uses artificial neural networks to process and analyse data, and can learn to recognize complex patterns in the environment [ 24 , 25 , 26 ].…”

Reflectance Spectroscopy for the Classification and Prediction of Pigments in Agronomic Crops

“…This advancement allows decision-makers to utilize computational intelligence to enhance the classification, monitoring, and nutritional value of crops in fields, resulting in billions of dollars in annual economic benefits [ 13 ]. Consequently, artificial intelligence algorithms (AIAs) based on data mining (DM) [ 18 ], deep learning (DL) [ 7 , 19 ], and machine learning (ML) [ 20 , 21 , 22 ] present promising techniques for future non-invasive pigment analyses in crop sciences and remote sensing applications.…”

Research on the Process of Convective Drying of Apples and Apricots Using an Original Drying Installation

“…In the past, the evaluation of the color of fruits was done subjectively, based on visual evaluation. For this purpose, a black box was designed and built for photometric determinations regarding the change in fruit color (due to oxidation processes) during to exposure at room temperature, according to the method described by Ropelewska [43] and Baigts-Allende et al [44]. Three varieties of apples were chosen: Golden Delicious, Jonagold and Idared; they were subjected to the oxidation process.…”