Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model

Aflatoxin B1, a toxic carcinogen frequently contaminating almonds, nuts, and food products, poses significant health risks. Therefore, a rapid and non-destructive detection method is crucial to detect aflatoxin B1-contaminated almonds to ensure food safety. This study introduces a novel deep learnin...

Full description

Saved in:
Bibliographic Details
Main Authors: Md. Ahasan Kabir, Ivan Lee, Sang-Heon Lee
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Toxins
Subjects:
Online Access:https://www.mdpi.com/2072-6651/17/4/156
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713301088370688
author Md. Ahasan Kabir
Ivan Lee
Sang-Heon Lee
author_facet Md. Ahasan Kabir
Ivan Lee
Sang-Heon Lee
author_sort Md. Ahasan Kabir
collection DOAJ
description Aflatoxin B1, a toxic carcinogen frequently contaminating almonds, nuts, and food products, poses significant health risks. Therefore, a rapid and non-destructive detection method is crucial to detect aflatoxin B1-contaminated almonds to ensure food safety. This study introduces a novel deep learning approach utilizing 3D Inception–ResNet architecture with fine-tuning to classify aflatoxin B1-contaminated almonds using hyperspectral images. The proposed model achieved higher classification accuracy than traditional methods, such as support vector machine (SVM), random forest (RF), quadratic discriminant analysis (QDA), and decision tree (DT), for classifying aflatoxin B1 contaminated almonds. A feature selection algorithm was employed to enhance processing efficiency and reduce spectral dimensionality while maintaining high classification accuracy. Experimental results demonstrate that the proposed 3D Inception–ResNet (Lightweight) model achieves superior classification performance with a 90.81% validation accuracy, an F1-score of 0.899, and an area under the curve value of 0.964, outperforming traditional machine learning approaches. The Lightweight 3D Inception–ResNet model, with 381 layers, offers a computationally efficient alternative suitable for real-time industrial applications. These research findings highlight the potential of hyperspectral imaging combined with deep learning for aflatoxin B1 detection in almonds with higher accuracy. This approach supports the development of real-time automated screening systems for food safety, reducing contamination-related risks in almonds.
format Article
id doaj-art-73d0eff6328c4998b0b19046788d087a
institution DOAJ
issn 2072-6651
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Toxins
spelling doaj-art-73d0eff6328c4998b0b19046788d087a2025-08-20T03:13:59ZengMDPI AGToxins2072-66512025-03-0117415610.3390/toxins17040156Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet ModelMd. Ahasan Kabir0Ivan Lee1Sang-Heon Lee2UniSA STEM, University of South Australia, Mawson Lakes, Adelaide, SA 5095, AustraliaUniSA STEM, University of South Australia, Mawson Lakes, Adelaide, SA 5095, AustraliaUniSA STEM, University of South Australia, Mawson Lakes, Adelaide, SA 5095, AustraliaAflatoxin B1, a toxic carcinogen frequently contaminating almonds, nuts, and food products, poses significant health risks. Therefore, a rapid and non-destructive detection method is crucial to detect aflatoxin B1-contaminated almonds to ensure food safety. This study introduces a novel deep learning approach utilizing 3D Inception–ResNet architecture with fine-tuning to classify aflatoxin B1-contaminated almonds using hyperspectral images. The proposed model achieved higher classification accuracy than traditional methods, such as support vector machine (SVM), random forest (RF), quadratic discriminant analysis (QDA), and decision tree (DT), for classifying aflatoxin B1 contaminated almonds. A feature selection algorithm was employed to enhance processing efficiency and reduce spectral dimensionality while maintaining high classification accuracy. Experimental results demonstrate that the proposed 3D Inception–ResNet (Lightweight) model achieves superior classification performance with a 90.81% validation accuracy, an F1-score of 0.899, and an area under the curve value of 0.964, outperforming traditional machine learning approaches. The Lightweight 3D Inception–ResNet model, with 381 layers, offers a computationally efficient alternative suitable for real-time industrial applications. These research findings highlight the potential of hyperspectral imaging combined with deep learning for aflatoxin B1 detection in almonds with higher accuracy. This approach supports the development of real-time automated screening systems for food safety, reducing contamination-related risks in almonds.https://www.mdpi.com/2072-6651/17/4/156aflatoxin B1hyperspectral imagingInception–ResNetconvolutional neural networkdeep learningAUC
spellingShingle Md. Ahasan Kabir
Ivan Lee
Sang-Heon Lee
Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
Toxins
aflatoxin B1
hyperspectral imaging
Inception–ResNet
convolutional neural network
deep learning
AUC
title Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
title_full Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
title_fullStr Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
title_full_unstemmed Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
title_short Deep Learning-Based Detection of Aflatoxin B1 Contamination in Almonds Using Hyperspectral Imaging: A Focus on Optimized 3D Inception–ResNet Model
title_sort deep learning based detection of aflatoxin b1 contamination in almonds using hyperspectral imaging a focus on optimized 3d inception resnet model
topic aflatoxin B1
hyperspectral imaging
Inception–ResNet
convolutional neural network
deep learning
AUC
url https://www.mdpi.com/2072-6651/17/4/156
work_keys_str_mv AT mdahasankabir deeplearningbaseddetectionofaflatoxinb1contaminationinalmondsusinghyperspectralimagingafocusonoptimized3dinceptionresnetmodel
AT ivanlee deeplearningbaseddetectionofaflatoxinb1contaminationinalmondsusinghyperspectralimagingafocusonoptimized3dinceptionresnetmodel
AT sangheonlee deeplearningbaseddetectionofaflatoxinb1contaminationinalmondsusinghyperspectralimagingafocusonoptimized3dinceptionresnetmodel