Cargando…

ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images

Wheat seed classification is a critical task for ensuring crop quality and yield. However, the characteristics of wheat seeds can vary due to variations in climate, soil, and other environmental factors across different years. Consequently, the present classification model is no longer adequate for...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhao, Xin, Liu, Shuo, Que, Haotian, Huang, Min, Zhu, Qibing
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10575222/
https://www.ncbi.nlm.nih.gov/pubmed/37836946
http://dx.doi.org/10.3390/s23198116
_version_ 1785120875074289664
author Zhao, Xin
Liu, Shuo
Que, Haotian
Huang, Min
Zhu, Qibing
author_facet Zhao, Xin
Liu, Shuo
Que, Haotian
Huang, Min
Zhu, Qibing
author_sort Zhao, Xin
collection PubMed
description Wheat seed classification is a critical task for ensuring crop quality and yield. However, the characteristics of wheat seeds can vary due to variations in climate, soil, and other environmental factors across different years. Consequently, the present classification model is no longer adequate for accurately classifying novel samples. To tackle this issue, this paper proposes an adaptive domain feature separation (ADFS) network that utilizes hyperspectral imaging techniques for cross-year classification of wheat seed varieties. The primary objective is to improve the generalization ability of the model at a minimum cost. ADFS leverages deep learning techniques to acquire domain-irrelevant features from hyperspectral data, thus effectively addressing the issue of domain shifts across datasets. The feature spaces are divided into three parts using different modules. One shared module aligns feature distributions between the source and target datasets from different years, thereby enhancing the model’s generalization and robustness. Additionally, two private modules extract class-specific features and domain-specific features. The transfer mechanism does not learn domain-specific features to reduce negative transfer and improve classification accuracy. Extensive experiments conducted on a two-year dataset comprising four wheat seed varieties demonstrate the effectiveness of ADFS in wheat seed classification. Compared with three typical transfer learning networks, ADFS can achieve the best accuracy of wheat seed classification with small batch samples updated, thereby addressing new seasonal variability.
format Online
Article
Text
id pubmed-10575222
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-105752222023-10-14 ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images Zhao, Xin Liu, Shuo Que, Haotian Huang, Min Zhu, Qibing Sensors (Basel) Article Wheat seed classification is a critical task for ensuring crop quality and yield. However, the characteristics of wheat seeds can vary due to variations in climate, soil, and other environmental factors across different years. Consequently, the present classification model is no longer adequate for accurately classifying novel samples. To tackle this issue, this paper proposes an adaptive domain feature separation (ADFS) network that utilizes hyperspectral imaging techniques for cross-year classification of wheat seed varieties. The primary objective is to improve the generalization ability of the model at a minimum cost. ADFS leverages deep learning techniques to acquire domain-irrelevant features from hyperspectral data, thus effectively addressing the issue of domain shifts across datasets. The feature spaces are divided into three parts using different modules. One shared module aligns feature distributions between the source and target datasets from different years, thereby enhancing the model’s generalization and robustness. Additionally, two private modules extract class-specific features and domain-specific features. The transfer mechanism does not learn domain-specific features to reduce negative transfer and improve classification accuracy. Extensive experiments conducted on a two-year dataset comprising four wheat seed varieties demonstrate the effectiveness of ADFS in wheat seed classification. Compared with three typical transfer learning networks, ADFS can achieve the best accuracy of wheat seed classification with small batch samples updated, thereby addressing new seasonal variability. MDPI 2023-09-27 /pmc/articles/PMC10575222/ /pubmed/37836946 http://dx.doi.org/10.3390/s23198116 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Zhao, Xin
Liu, Shuo
Que, Haotian
Huang, Min
Zhu, Qibing
ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title_full ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title_fullStr ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title_full_unstemmed ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title_short ADFSNet: An Adaptive Domain Feature Separation Network for the Classification of Wheat Seed Using Hyperspectral Images
title_sort adfsnet: an adaptive domain feature separation network for the classification of wheat seed using hyperspectral images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10575222/
https://www.ncbi.nlm.nih.gov/pubmed/37836946
http://dx.doi.org/10.3390/s23198116
work_keys_str_mv AT zhaoxin adfsnetanadaptivedomainfeatureseparationnetworkfortheclassificationofwheatseedusinghyperspectralimages
AT liushuo adfsnetanadaptivedomainfeatureseparationnetworkfortheclassificationofwheatseedusinghyperspectralimages
AT quehaotian adfsnetanadaptivedomainfeatureseparationnetworkfortheclassificationofwheatseedusinghyperspectralimages
AT huangmin adfsnetanadaptivedomainfeatureseparationnetworkfortheclassificationofwheatseedusinghyperspectralimages
AT zhuqibing adfsnetanadaptivedomainfeatureseparationnetworkfortheclassificationofwheatseedusinghyperspectralimages