CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases
Grape disease is a significant contributory factor to the decline in grape yield, typically affecting the leaves first. Efficient identification of grape leaf diseases remains a critical unmet need. To mitigate background interference in grape leaf feature extraction and improve the ability to extra...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-05-01
|
Series: | Frontiers in Plant Science |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2022.846767/full |
_version_ | 1811255816238399488 |
---|---|
author | Jiayu Suo Jialei Zhan Guoxiong Zhou Aibin Chen Yaowen Hu Weiqi Huang Weiwei Cai Yahui Hu Liujun Li |
author_facet | Jiayu Suo Jialei Zhan Guoxiong Zhou Aibin Chen Yaowen Hu Weiqi Huang Weiwei Cai Yahui Hu Liujun Li |
author_sort | Jiayu Suo |
collection | DOAJ |
description | Grape disease is a significant contributory factor to the decline in grape yield, typically affecting the leaves first. Efficient identification of grape leaf diseases remains a critical unmet need. To mitigate background interference in grape leaf feature extraction and improve the ability to extract small disease spots, by combining the characteristic features of grape leaf diseases, we developed a novel method for disease recognition and classification in this study. First, Gaussian filters Sobel smooth de-noising Laplace operator (GSSL) was employed to reduce image noise and enhance the texture of grape leaves. A novel network designated coordinated attention shuffle mechanism-asymmetric multi-scale fusion module net (CASM-AMFMNet) was subsequently applied for grape leaf disease identification. CoAtNet was employed as the network backbone to improve model learning and generalization capabilities, which alleviated the problem of gradient explosion to a certain extent. The CASM-AMFMNet was further utilized to capture and target grape leaf disease areas, therefore reducing background interference. Finally, Asymmetric multi-scale fusion module (AMFM) was employed to extract multi-scale features from small disease spots on grape leaves for accurate identification of small target diseases. The experimental results based on our self-made grape leaf image dataset showed that, compared to existing methods, CASM-AMFMNet achieved an accuracy of 95.95%, F1 score of 95.78%, and mAP of 90.27%. Overall, the model and methods proposed in this report could successfully identify different diseases of grape leaves and provide a feasible scheme for deep learning to correctly recognize grape diseases during agricultural production that may be used as a reference for other crops diseases. |
first_indexed | 2024-04-12T17:29:42Z |
format | Article |
id | doaj.art-fbf1a36ffea741efaab6adc77029adbd |
institution | Directory Open Access Journal |
issn | 1664-462X |
language | English |
last_indexed | 2024-04-12T17:29:42Z |
publishDate | 2022-05-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Plant Science |
spelling | doaj.art-fbf1a36ffea741efaab6adc77029adbd2022-12-22T03:23:10ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2022-05-011310.3389/fpls.2022.846767846767CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf DiseasesJiayu Suo0Jialei Zhan1Guoxiong Zhou2Aibin Chen3Yaowen Hu4Weiqi Huang5Weiwei Cai6Yahui Hu7Liujun Li8College of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaCollege of Computer and Information Engineering, Central South University of Forestry and Technology, Changsha, ChinaPlant Protection Research Institute, Hunan Academy of Agricultural Sciences (HNAAS), Changsha, ChinaDepartment of Civil, Architectural and Environmental Engineering, Missouri University of Science and Technology, Rolla, MO, United StatesGrape disease is a significant contributory factor to the decline in grape yield, typically affecting the leaves first. Efficient identification of grape leaf diseases remains a critical unmet need. To mitigate background interference in grape leaf feature extraction and improve the ability to extract small disease spots, by combining the characteristic features of grape leaf diseases, we developed a novel method for disease recognition and classification in this study. First, Gaussian filters Sobel smooth de-noising Laplace operator (GSSL) was employed to reduce image noise and enhance the texture of grape leaves. A novel network designated coordinated attention shuffle mechanism-asymmetric multi-scale fusion module net (CASM-AMFMNet) was subsequently applied for grape leaf disease identification. CoAtNet was employed as the network backbone to improve model learning and generalization capabilities, which alleviated the problem of gradient explosion to a certain extent. The CASM-AMFMNet was further utilized to capture and target grape leaf disease areas, therefore reducing background interference. Finally, Asymmetric multi-scale fusion module (AMFM) was employed to extract multi-scale features from small disease spots on grape leaves for accurate identification of small target diseases. The experimental results based on our self-made grape leaf image dataset showed that, compared to existing methods, CASM-AMFMNet achieved an accuracy of 95.95%, F1 score of 95.78%, and mAP of 90.27%. Overall, the model and methods proposed in this report could successfully identify different diseases of grape leaves and provide a feasible scheme for deep learning to correctly recognize grape diseases during agricultural production that may be used as a reference for other crops diseases.https://www.frontiersin.org/articles/10.3389/fpls.2022.846767/fullCASM-AMFMNetcoordinate attention shuffle mechanism asymmetricmulti-scale fusion modulegrape leaf diseasesGSSLimage enhancement |
spellingShingle | Jiayu Suo Jialei Zhan Guoxiong Zhou Aibin Chen Yaowen Hu Weiqi Huang Weiwei Cai Yahui Hu Liujun Li CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases Frontiers in Plant Science CASM-AMFMNet coordinate attention shuffle mechanism asymmetric multi-scale fusion module grape leaf diseases GSSL image enhancement |
title | CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases |
title_full | CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases |
title_fullStr | CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases |
title_full_unstemmed | CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases |
title_short | CASM-AMFMNet: A Network Based on Coordinate Attention Shuffle Mechanism and Asymmetric Multi-Scale Fusion Module for Classification of Grape Leaf Diseases |
title_sort | casm amfmnet a network based on coordinate attention shuffle mechanism and asymmetric multi scale fusion module for classification of grape leaf diseases |
topic | CASM-AMFMNet coordinate attention shuffle mechanism asymmetric multi-scale fusion module grape leaf diseases GSSL image enhancement |
url | https://www.frontiersin.org/articles/10.3389/fpls.2022.846767/full |
work_keys_str_mv | AT jiayusuo casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT jialeizhan casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT guoxiongzhou casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT aibinchen casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT yaowenhu casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT weiqihuang casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT weiweicai casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT yahuihu casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases AT liujunli casmamfmnetanetworkbasedoncoordinateattentionshufflemechanismandasymmetricmultiscalefusionmoduleforclassificationofgrapeleafdiseases |