Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings

The agro-industrial sector faces significant challenges in product classification, which directly affect product quality, production efficiency and food safety. This paper proposes a machine learning model that correctly identifies the different attributes of Persea americana. For this, an automatic...

Full description

Saved in:
Bibliographic Details
Main Authors: Oscar Vera, Jose Cruz, Severo Huaquipaco, Wilson Mamani, Victor Yana-Mamani, Saul Huaquipaco
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10750816/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849220017509367808
author Oscar Vera
Jose Cruz
Severo Huaquipaco
Wilson Mamani
Victor Yana-Mamani
Saul Huaquipaco
author_facet Oscar Vera
Jose Cruz
Severo Huaquipaco
Wilson Mamani
Victor Yana-Mamani
Saul Huaquipaco
author_sort Oscar Vera
collection DOAJ
description The agro-industrial sector faces significant challenges in product classification, which directly affect product quality, production efficiency and food safety. This paper proposes a machine learning model that correctly identifies the different attributes of Persea americana. For this, an automatic agro-industrial plant was implemented following industrial standards where advanced image processing techniques were used on a dataset of 346 images for training and 146 images for testing, with three deep convolutional neural networks with improved training strategies and advanced validation techniques including True Skill Statistic (TSS), Cohen’s Kappa (K), Threat Score (TS), Heidke Skill Score (HSS) and Probability of Error (Pe). The results showed that the DenseNet model outperforms other state-of-the-art models in accuracy, reaching an F1 score of 99.27%, ResNet 50 reached 99.26% and EfficientNet B4 reached 99.19%, also in the validation phase TSS, K, TS and HSS for all models were higher than 0.98 while the Pe index was higher than 0.55. It is concluded that the DenseNet model is shown to be the effective and reliable technique for the classification of Persea Americana. These promising results open new possibilities for the implementation of machine learning in the agri-food industry. For future research, the possibility of expanding the data set and extending the application of this model to other varieties of fruits is proposed.
format Article
id doaj-art-76cd1569b7194d39bad57425e11b3d33
institution Kabale University
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-76cd1569b7194d39bad57425e11b3d332024-12-25T00:01:23ZengIEEEIEEE Access2169-35362024-01-011219424019425510.1109/ACCESS.2024.349672810750816Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial SettingsOscar Vera0https://orcid.org/0000-0002-1996-8471Jose Cruz1https://orcid.org/0000-0002-5201-0265Severo Huaquipaco2Wilson Mamani3Victor Yana-Mamani4https://orcid.org/0000-0003-0982-2353Saul Huaquipaco5https://orcid.org/0000-0003-2323-3061School of Ingeniería de Sistemas e Informatica, Faculty of Engineering, Universidad Nacional de Moquegua, Moquegua, PeruSchool of Ingeniería Electrónica, Faculty of Mecánica Eléctrica, Electrónica y Sistemas, Universidad Nacional del Altiplano de Puno, Puno, PeruSchool of Ingeniería Agroindustrial, Faculty of Ciencias Agrarias, Universidad Nacional de Huancavelica, Paturpampa, PeruUniversity de Alicante - EspaÃa, Alicante, SpainSchool of Ingeniería de Sistemas e Informatica, Faculty of Engineering, Universidad Nacional de Moquegua, Moquegua, PeruSchool of Ingeniería de Sistemas e Informatica, Faculty of Engineering, Universidad Nacional de Moquegua, Moquegua, PeruThe agro-industrial sector faces significant challenges in product classification, which directly affect product quality, production efficiency and food safety. This paper proposes a machine learning model that correctly identifies the different attributes of Persea americana. For this, an automatic agro-industrial plant was implemented following industrial standards where advanced image processing techniques were used on a dataset of 346 images for training and 146 images for testing, with three deep convolutional neural networks with improved training strategies and advanced validation techniques including True Skill Statistic (TSS), Cohen’s Kappa (K), Threat Score (TS), Heidke Skill Score (HSS) and Probability of Error (Pe). The results showed that the DenseNet model outperforms other state-of-the-art models in accuracy, reaching an F1 score of 99.27%, ResNet 50 reached 99.26% and EfficientNet B4 reached 99.19%, also in the validation phase TSS, K, TS and HSS for all models were higher than 0.98 while the Pe index was higher than 0.55. It is concluded that the DenseNet model is shown to be the effective and reliable technique for the classification of Persea Americana. These promising results open new possibilities for the implementation of machine learning in the agri-food industry. For future research, the possibility of expanding the data set and extending the application of this model to other varieties of fruits is proposed.https://ieeexplore.ieee.org/document/10750816/Classification Persea Americanaconvolutional neuronal networkDenseNetEfficientNetResNet
spellingShingle Oscar Vera
Jose Cruz
Severo Huaquipaco
Wilson Mamani
Victor Yana-Mamani
Saul Huaquipaco
Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
IEEE Access
Classification Persea Americana
convolutional neuronal network
DenseNet
EfficientNet
ResNet
title Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
title_full Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
title_fullStr Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
title_full_unstemmed Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
title_short Automated and Enhanced Classification of Persea Americana Using Optimized Deep Convolutional Neural Networks With Improved Training Strategies for Agro-Industrial Settings
title_sort automated and enhanced classification of persea americana using optimized deep convolutional neural networks with improved training strategies for agro industrial settings
topic Classification Persea Americana
convolutional neuronal network
DenseNet
EfficientNet
ResNet
url https://ieeexplore.ieee.org/document/10750816/
work_keys_str_mv AT oscarvera automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings
AT josecruz automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings
AT severohuaquipaco automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings
AT wilsonmamani automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings
AT victoryanamamani automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings
AT saulhuaquipaco automatedandenhancedclassificationofperseaamericanausingoptimizeddeepconvolutionalneuralnetworkswithimprovedtrainingstrategiesforagroindustrialsettings