Cargando…

Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition

Deep learning shows its advantages and potentials in plant disease recognition and has witnessed a profound development in recent years. To obtain a competing performance with a deep learning algorithm, enough amount of annotated data is requested but in the natural world, scarce or imbalanced data...

Descripción completa

Detalles Bibliográficos
Autores principales: Xu, Mingle, Yoon, Sook, Fuentes, Alvaro, Yang, Jucheng, Park, Dong Sun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8858820/
https://www.ncbi.nlm.nih.gov/pubmed/35197989
http://dx.doi.org/10.3389/fpls.2021.773142
_version_ 1784654319487811584
author Xu, Mingle
Yoon, Sook
Fuentes, Alvaro
Yang, Jucheng
Park, Dong Sun
author_facet Xu, Mingle
Yoon, Sook
Fuentes, Alvaro
Yang, Jucheng
Park, Dong Sun
author_sort Xu, Mingle
collection PubMed
description Deep learning shows its advantages and potentials in plant disease recognition and has witnessed a profound development in recent years. To obtain a competing performance with a deep learning algorithm, enough amount of annotated data is requested but in the natural world, scarce or imbalanced data are common, and annotated data is expensive or hard to collect. Data augmentation, aiming to create variations for training data, has shown its power for this issue. But there are still two challenges: creating more desirable variations for scarce and imbalanced data, and designing a data augmentation to ease object detection and instance segmentation. First, current algorithms made variations only inside one specific class, but more desirable variations can further promote performance. To address this issue, we propose a novel data augmentation paradigm that can adapt variations from one class to another. In the novel paradigm, an image in the source domain is translated into the target domain, while the variations unrelated to the domain are maintained. For example, an image with a healthy tomato leaf is translated into a powdery mildew image but the variations of the healthy leaf are maintained and transferred into the powdery mildew class, such as types of tomato leaf, sizes, and viewpoints. Second, current data augmentation is suitable to promote the image classification model but may not be appropriate to alleviate object detection and instance segmentation model, mainly because the necessary annotations can not be obtained. In this study, we leverage a prior mask as input to tell the area we are interested in and reuse the original annotations. In this way, our proposed algorithm can be utilized to do the three tasks simultaneously. Further, We collect 1,258 images of tomato leaves with 1,429 instance segmentation annotations as there is more than one instance in one single image, including five diseases and healthy leaves. Extensive experimental results on the collected images validate that our new data augmentation algorithm makes useful variations and contributes to improving performance for diverse deep learning-based methods.
format Online
Article
Text
id pubmed-8858820
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-88588202022-02-22 Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition Xu, Mingle Yoon, Sook Fuentes, Alvaro Yang, Jucheng Park, Dong Sun Front Plant Sci Plant Science Deep learning shows its advantages and potentials in plant disease recognition and has witnessed a profound development in recent years. To obtain a competing performance with a deep learning algorithm, enough amount of annotated data is requested but in the natural world, scarce or imbalanced data are common, and annotated data is expensive or hard to collect. Data augmentation, aiming to create variations for training data, has shown its power for this issue. But there are still two challenges: creating more desirable variations for scarce and imbalanced data, and designing a data augmentation to ease object detection and instance segmentation. First, current algorithms made variations only inside one specific class, but more desirable variations can further promote performance. To address this issue, we propose a novel data augmentation paradigm that can adapt variations from one class to another. In the novel paradigm, an image in the source domain is translated into the target domain, while the variations unrelated to the domain are maintained. For example, an image with a healthy tomato leaf is translated into a powdery mildew image but the variations of the healthy leaf are maintained and transferred into the powdery mildew class, such as types of tomato leaf, sizes, and viewpoints. Second, current data augmentation is suitable to promote the image classification model but may not be appropriate to alleviate object detection and instance segmentation model, mainly because the necessary annotations can not be obtained. In this study, we leverage a prior mask as input to tell the area we are interested in and reuse the original annotations. In this way, our proposed algorithm can be utilized to do the three tasks simultaneously. Further, We collect 1,258 images of tomato leaves with 1,429 instance segmentation annotations as there is more than one instance in one single image, including five diseases and healthy leaves. Extensive experimental results on the collected images validate that our new data augmentation algorithm makes useful variations and contributes to improving performance for diverse deep learning-based methods. Frontiers Media S.A. 2022-02-07 /pmc/articles/PMC8858820/ /pubmed/35197989 http://dx.doi.org/10.3389/fpls.2021.773142 Text en Copyright © 2022 Xu, Yoon, Fuentes, Yang and Park. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Plant Science
Xu, Mingle
Yoon, Sook
Fuentes, Alvaro
Yang, Jucheng
Park, Dong Sun
Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title_full Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title_fullStr Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title_full_unstemmed Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title_short Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition
title_sort style-consistent image translation: a novel data augmentation paradigm to improve plant disease recognition
topic Plant Science
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8858820/
https://www.ncbi.nlm.nih.gov/pubmed/35197989
http://dx.doi.org/10.3389/fpls.2021.773142
work_keys_str_mv AT xumingle styleconsistentimagetranslationanoveldataaugmentationparadigmtoimproveplantdiseaserecognition
AT yoonsook styleconsistentimagetranslationanoveldataaugmentationparadigmtoimproveplantdiseaserecognition
AT fuentesalvaro styleconsistentimagetranslationanoveldataaugmentationparadigmtoimproveplantdiseaserecognition
AT yangjucheng styleconsistentimagetranslationanoveldataaugmentationparadigmtoimproveplantdiseaserecognition
AT parkdongsun styleconsistentimagetranslationanoveldataaugmentationparadigmtoimproveplantdiseaserecognition