Machine Learning for Automatic Classification of Tomato Ripening Stages Using Electrophysiological Recordings

Abstract

The physiological processes underlying fruit ripening can lead to different electrical signatures at each ripening stage, making it possible to classify tomato fruit through the analysis of electrical signals. Here, the electrical activity of tomato fruit (Solanum lycopersicum var. cerasiforme) during ripening was investigated as tissue voltage variations, and Machine Learning (ML) techniques were used for the classification of different ripening stages. Tomato fruit was harvested at the mature green stage and placed in a Faraday's cage under laboratory-controlled conditions. Two electrodes per fruit were inserted 1 cm apart from each other. The measures were carried out continuously until the entire fruits reached the light red stage. The time series were analyzed by the following techniques: Fast Fourier Transform (FFT), Wavelet Transform, Power Spectral Density (PSD), and Approximate Entropy. Descriptive analysis from FFT, PSD, and Wavelet Transform were used for PCA (Principal Component Analysis). Finally, ApEn, PCA1, PCA2, and PCA3 were obtained. These features were used in ML analyses for looking for classifiable patterns of the three different ripening stages: mature green, breaker, and light red. The results showed that it is possible to classify the ripening stages using the fruit's electrical activity. It was also observed, using precision, sensitivity, and F1-score techniques, that the breaker stage was the most classifiable among all stages. It was found a more accurate distinction between mature green × breaker than between breaker × light red. The ML techniques used seem to be a novel tool for classifying ripening stages. The features obtained from electrophysiological time series have the potential to be used for supervised training, being able to help in more accurate classification of fruit ripening stages.

Systemic Signals Induced by Single and Combined Abiotic Stimuli in Common Bean Plants

“…In addition, each stimulus can affect the plant in a specific way and trigger different responses in each module [ 3 ]. Therefore, the focus of recent research has been understanding how plants respond to different stimuli, which signals are activated, and if there is a pattern that can be identified [ 15 , 16 , 17 , 18 , 27 ].…”

“…The TDAF method, developed by our research group, aims to demonstrate the dynamics of the electrome run characteristics (features). The latest studies published in the literature suggest that 3 min long measurements already contain enough information to identify behaviors in the plant electrome [ 16 , 17 , 18 ]. Thus, the TDAF assumes the use of the smallest possible unit of the analyzed series, considering the measurement equipment, and uses the dispersion theory [ 51 ] to analyze the dataset as a whole.…”

“…Recent studies analyzing plant electrome [ 1 , 13 , 14 , 15 , 16 , 17 ] revealed its potential as a tool for early stress diagnosis in plants since the electrome exhibits specific patterns of responses that function as a classifiable electrical signature by standard time series analysis and machine learning techniques [ 15 , 16 , 17 , 18 , 19 ].…”

Approximate entropy: a promising tool to understand the hidden electrical activity of fruit

“…Recently, our research group has provided evidence that supports [ 10 ], study. We found classifiable differences in the electrome of tomato fruits during ripening [ 43 ], as well as systemic electrical signaling from fruits being chewed by caterpillars toward the rest of the aerial part of the plant (Reissig et al, 2021a). These changes in the electrome are likely associated with physiological and cognitive processes [ 11–13 ].…”

“…We have previously published works demonstrating that signal complexity analyses, such as approximate entropy (ApEn), provided important features for Machine Learning (ML) classification of the fruit electrome changes (Reissig et al, 2021a, 43 ). Now, we bring some unpublished data, obtained from the experiment conducted by [ 15 ], that is part of larger effort to understand these signals in fruits.…”