AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases

Soybean is a critical agricultural commodity, serving as a vital source of protein and vegetable oil, and contributing significantly to the economies of producing nations. However, soybean yields are frequently compromised by disease and pest infestations, which, if not identified early, can lead to...

Full description

Saved in:
Bibliographic Details
Main Authors: Oluwatoyin Joy Omole, Renata Lopes Rosa, Muhammad Saadi, Demóstenes Zegarra Rodriguez
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:AI
Subjects:
Online Access:https://www.mdpi.com/2673-2688/5/4/142
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846106288542449664
author Oluwatoyin Joy Omole
Renata Lopes Rosa
Muhammad Saadi
Demóstenes Zegarra Rodriguez
author_facet Oluwatoyin Joy Omole
Renata Lopes Rosa
Muhammad Saadi
Demóstenes Zegarra Rodriguez
author_sort Oluwatoyin Joy Omole
collection DOAJ
description Soybean is a critical agricultural commodity, serving as a vital source of protein and vegetable oil, and contributing significantly to the economies of producing nations. However, soybean yields are frequently compromised by disease and pest infestations, which, if not identified early, can lead to substantial production losses. To address this challenge, we propose AgriNAS, a method that integrates a Neural Architecture Search (NAS) framework with an adaptive convolutional architecture specifically designed for plant pathology. AgriNAS employs a novel data augmentation strategy and a Spatial–Time Augmentation (STA) method, and it utilizes a multi-stage convolutional network that dynamically adapts to the complexity of the input data. The proposed AgriNAS leverages powerful GPU resources to handle the intensive computational tasks involved in NAS and model training. The framework incorporates a bi-level optimization strategy and entropy-based regularization to enhance model robustness and prevent overfitting. AgriNAS achieves classification accuracies superior to VGG-19 and a transfer learning method using convolutional neural networks.
format Article
id doaj-art-9ae6ce37aa1340bf969e6a94a90f2d5d
institution Kabale University
issn 2673-2688
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series AI
spelling doaj-art-9ae6ce37aa1340bf969e6a94a90f2d5d2024-12-27T14:05:08ZengMDPI AGAI2673-26882024-12-01542945296610.3390/ai5040142AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean DiseasesOluwatoyin Joy Omole0Renata Lopes Rosa1Muhammad Saadi2Demóstenes Zegarra Rodriguez3Department of Computer Science, Federal University of Lavras, Lavras 37200-000, Minas Gerais, BrazilDepartment of Computer Science, Federal University of Lavras, Lavras 37200-000, Minas Gerais, BrazilSchool of Science and Technology, Department of Computer Science, Nottingham Trent University, Nottingham NG11 8NS , UKDepartment of Computer Science, Federal University of Lavras, Lavras 37200-000, Minas Gerais, BrazilSoybean is a critical agricultural commodity, serving as a vital source of protein and vegetable oil, and contributing significantly to the economies of producing nations. However, soybean yields are frequently compromised by disease and pest infestations, which, if not identified early, can lead to substantial production losses. To address this challenge, we propose AgriNAS, a method that integrates a Neural Architecture Search (NAS) framework with an adaptive convolutional architecture specifically designed for plant pathology. AgriNAS employs a novel data augmentation strategy and a Spatial–Time Augmentation (STA) method, and it utilizes a multi-stage convolutional network that dynamically adapts to the complexity of the input data. The proposed AgriNAS leverages powerful GPU resources to handle the intensive computational tasks involved in NAS and model training. The framework incorporates a bi-level optimization strategy and entropy-based regularization to enhance model robustness and prevent overfitting. AgriNAS achieves classification accuracies superior to VGG-19 and a transfer learning method using convolutional neural networks.https://www.mdpi.com/2673-2688/5/4/142neural architecture searchadaptive convolutional architecturedata augmentationsoybean disease and pest detection
spellingShingle Oluwatoyin Joy Omole
Renata Lopes Rosa
Muhammad Saadi
Demóstenes Zegarra Rodriguez
AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
AI
neural architecture search
adaptive convolutional architecture
data augmentation
soybean disease and pest detection
title AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
title_full AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
title_fullStr AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
title_full_unstemmed AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
title_short AgriNAS: Neural Architecture Search with Adaptive Convolution and Spatial–Time Augmentation Method for Soybean Diseases
title_sort agrinas neural architecture search with adaptive convolution and spatial time augmentation method for soybean diseases
topic neural architecture search
adaptive convolutional architecture
data augmentation
soybean disease and pest detection
url https://www.mdpi.com/2673-2688/5/4/142
work_keys_str_mv AT oluwatoyinjoyomole agrinasneuralarchitecturesearchwithadaptiveconvolutionandspatialtimeaugmentationmethodforsoybeandiseases
AT renatalopesrosa agrinasneuralarchitecturesearchwithadaptiveconvolutionandspatialtimeaugmentationmethodforsoybeandiseases
AT muhammadsaadi agrinasneuralarchitecturesearchwithadaptiveconvolutionandspatialtimeaugmentationmethodforsoybeandiseases
AT demosteneszegarrarodriguez agrinasneuralarchitecturesearchwithadaptiveconvolutionandspatialtimeaugmentationmethodforsoybeandiseases