Freeze-Damage Detection in Lemons Using Electrochemical Impedance Spectroscopy

Fernández, Adrián Ochandio; Pinatti, Cristian Ariel Olguín; Peris, Rafael Masot; Laguarda-Miró, N.

doi:10.3390/s19184051

Cited by 23 publications

(13 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to previously mentioned works, presenting a 90% precision in the determination of avocado post-harvest ripening degree 28 using ANN and 100% in the discrimination of frozen lemons using SVM 29 , the on-plant discrimination of strawberry ripening degree (0.72 F -score, MLP-1) resulted to be less effective. On the other hand, such works were carried out with a smaller number of fruit samples (respectively 100 and 180, compared to the 923 used in this work) and do not indicate if the test set was left unseen until the final classification task or was used in the training.…”

Section: Resultsmentioning

confidence: 56%

“…Nevertheless, all these works lack of well-established machine learning methods predicting the fruit ripening stage starting from bioimpedance data. The most relevant works in this specific field describe the application of support-vector machine algorithms to the classification of avocado ripening degree 28 and the detection of freeze damage in lemons using artificial neutral networks and principal component analysis 29 . Both papers achieved a good accuracy in the classification task, obtaining a 90% (SVM) and 100% (ANN) precision, training the models with a total 100 and 180 samples, respectively.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data

Ibba

Tronstad

Moscetti

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Strawberry is one of the most popular fruits in the market. To meet the demanding consumer and market quality standards, there is a strong need for an on-site, accurate and reliable grading system during the whole harvesting process. In this work, a total of 923 strawberry fruit were measured directly on-plant at different ripening stages by means of bioimpedance data, collected at frequencies between 20 Hz and 300 kHz. The fruit batch was then splitted in 2 classes (i.e. ripe and unripe) based on surface color data. Starting from these data, six of the most commonly used supervised machine learning classification techniques, i.e. Logistic Regression (LR), Binary Decision Trees (DT), Naive Bayes Classifiers (NBC), K-Nearest Neighbors (KNN), Support Vector Machine (SVM) and Multi-Layer Perceptron Networks (MLP), were employed, optimized, tested and compared in view of their performance in predicting the strawberry fruit ripening stage. Such models were trained to develop a complete feature selection and optimization pipeline, not yet available for bioimpedance data analysis of fruit. The classification results highlighted that, among all the tested methods, MLP networks had the best performances on the test set, with 0.72, 0.82 and 0.73 for the F$$_1$$ 1 , F$$_{0.5}$$ 0.5 and F$$_2$$ 2 -score, respectively, and improved the training results, showing good generalization capability, adapting well to new, previously unseen data. Consequently, the MLP models, trained with bioimpedance data, are a promising alternative for real-time estimation of strawberry ripeness directly on-field, which could be a potential application technique for evaluating the harvesting time management for farmers and producers.

show abstract

Section: Resultsmentioning

confidence: 56%

Section: Introductionmentioning

confidence: 99%

Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data

Ibba

Tronstad

Moscetti

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…(5), where Z j j is the modulus of impedance and θ is the phase angle. Previous studies have reported that freezing level can be identified through impedance parameters (Aboalnaga, Said, Madian, & Radwan, 2019;Fernández, Pinatti, Peris, & Laguarda-Miró, 2019;Liu et al, 2017;Ohnishi, Fujii, & Miyawaki, 2002;Vidaček, Medi c, Botka-Petrak, Nežak, & Petrak, 2008). These parameters, such as resistance and reactance, have an imperfect ability to investigate internal changes of a tissue.…”

Section: Introductionmentioning

confidence: 99%

A noninvasive method for detecting frozen injuries in potatoes based on electrical impedance spectroscopy

Feng

Hou

Zhang

2021

J Food Process Engineering

View full text Add to dashboard Cite

To simplify the application of electrical impedance spectroscopy, the characteristic points were explored for noninvasive determination of the freezing damage in potatoes. Impedance experiments were conducted on the potato samples at room temperature (20°C), 0, −4, and −20°C. Samples of fresh, frozen, and freeze–thawed samples were analyzed. The results show that the shapes of the impedance spectroscopies were different in each of these samples. The characteristic points Zextra, Zintr, and Zcyt from electrical impedance spectroscopy are proposed to indicate the freezing damage level. The Zcyt value was considerably different for each state (1,939; 25,130; and 43 Ω, respectively). This indicates that Zcyt is a useful index for freezing damage classification and that Zcyt could indicate cell structure changes in thawed samples. Practical applications The morphological characteristic points of electrical impedance spectroscopy were used to evaluate the freezing injury of potatoes. This method provides a novel approach to study the changes in biological tissues. The method of using characteristic points is more convenient and effective than with an equivalent circuit. The Zcyt value could be used for detecting freezing injury in potatoes. Moreover, this method can be potentially applied for quality analysis of agricultural products.

show abstract

“…Nevertheless, all these works lack of well-established machine learning methods predicting the fruit ripening stage starting from bioimpedance data. The most relevant works in this specific field describe the application of support-vector machine algorithms to the classification of avocado ripening degree 28 and the detection of freeze damage in lemons using artificial neutral networks and principal component analysis 29 . These works, despite being a good starting point for the use of a combined bioimpedance and machine learning approach for the evaluation of fruit quality, lack in the amount of considered data to develop the models and most importantly do not provide a detailed data analysis pipeline, which is strongly needed as a reference in the development of similar works.…”

Section: Introductionmentioning

confidence: 99%

Supervised Binary Classification Methods for Strawberry Ripeness Discrimination from Bioimpedance Data

Ibba

Tronstad

Moscetti

et al. 2020

Preprint

View full text Add to dashboard Cite

Strawberry is one of the most popular fruits in the market. To meet the demanding consumer and market quality standards, there is a strong need for an on-site, accurate and reliable grading system during the whole harvesting process. In this work, a total of 923 strawberry fruit were measured directly on-plant at different ripening stages by means of bioimpedance data, collected at frequencies between 20Hz and 300kHz. The fruit batch was then splitted in 2 classes (i.e. ripe and unripe) based on surface color data. Starting from these data, six of the most commonly used supervised machine learning classification techniques, i.e. Logistic Regression (LR), Binary Decision Trees (DT), Naive Bayes Classifiers (NBC), K-NearestNeighbors (KNN), Support Vector Machine (SVM) and Multi-Layer Perceptron Networks (MLP), were employed, optimized, tested and compared in view of their performance in predicting the strawberry fruit ripening stage. Such models were trained to develop a complete feature selection and optimization pipeline, not yet available for bioimpedance data analysis of fruit. The classification results highlighted that, among all the tested methods, MLP networks had the best performances on the test set, with 0.72, 0.82 and 0.73 for the F1, F0.5 and F2-score, respectively, and improved the training results, showing good generalization capability, adapting well to new, previously unseen data. Consequently, the MLP models, trained with bioimpedance data, are a promising alternative for real–time estimation of strawberry ripeness directly on–field, which could be a potential application technique for evaluating the harvesting time management for farmers and producers.

show abstract

Freeze-Damage Detection in Lemons Using Electrochemical Impedance Spectroscopy

Cited by 23 publications

References 43 publications

Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data

Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data

A noninvasive method for detecting frozen injuries in potatoes based on electrical impedance spectroscopy

Supervised Binary Classification Methods for Strawberry Ripeness Discrimination from Bioimpedance Data

Contact Info

Product

Resources

About