Cargando…

Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands

Recent developments in deep learning-based automatic weeding systems have shown promise for unmanned weed eradication. However, accurately distinguishing between crops and weeds in varying field conditions remains a challenge for these systems, as performance deteriorates when applied to new or diff...

Descripción completa

Detalles Bibliográficos
Autores principales: Ilyas, Talha, Lee, Jonghoon, Won, Okjae, Jeong, Yongchae, Kim, Hyongsuk
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10445656/
https://www.ncbi.nlm.nih.gov/pubmed/37621880
http://dx.doi.org/10.3389/fpls.2023.1234616
_version_ 1785094222424047616
author Ilyas, Talha
Lee, Jonghoon
Won, Okjae
Jeong, Yongchae
Kim, Hyongsuk
author_facet Ilyas, Talha
Lee, Jonghoon
Won, Okjae
Jeong, Yongchae
Kim, Hyongsuk
author_sort Ilyas, Talha
collection PubMed
description Recent developments in deep learning-based automatic weeding systems have shown promise for unmanned weed eradication. However, accurately distinguishing between crops and weeds in varying field conditions remains a challenge for these systems, as performance deteriorates when applied to new or different fields due to insignificant changes in low-level statistics and a significant gap between training and test data distributions. In this study, we propose an approach based on unsupervised domain adaptation to improve crop-weed recognition in new, unseen fields. Our system addresses this issue by learning to ignore insignificant changes in low-level statistics that cause a decline in performance when applied to new data. The proposed network includes a segmentation module that produces segmentation maps using labeled (training field) data while also minimizing entropy using unlabeled (test field) data simultaneously, and a discriminator module that maximizes the confusion between extracted features from the training and test farm samples. This module uses adversarial optimization to make the segmentation network invariant to changes in the field environment. We evaluated the proposed approach on four different unseen (test) fields and found consistent improvements in performance. These results suggest that the proposed approach can effectively handle changes in new field environments during real field inference.
format Online
Article
Text
id pubmed-10445656
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-104456562023-08-24 Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands Ilyas, Talha Lee, Jonghoon Won, Okjae Jeong, Yongchae Kim, Hyongsuk Front Plant Sci Plant Science Recent developments in deep learning-based automatic weeding systems have shown promise for unmanned weed eradication. However, accurately distinguishing between crops and weeds in varying field conditions remains a challenge for these systems, as performance deteriorates when applied to new or different fields due to insignificant changes in low-level statistics and a significant gap between training and test data distributions. In this study, we propose an approach based on unsupervised domain adaptation to improve crop-weed recognition in new, unseen fields. Our system addresses this issue by learning to ignore insignificant changes in low-level statistics that cause a decline in performance when applied to new data. The proposed network includes a segmentation module that produces segmentation maps using labeled (training field) data while also minimizing entropy using unlabeled (test field) data simultaneously, and a discriminator module that maximizes the confusion between extracted features from the training and test farm samples. This module uses adversarial optimization to make the segmentation network invariant to changes in the field environment. We evaluated the proposed approach on four different unseen (test) fields and found consistent improvements in performance. These results suggest that the proposed approach can effectively handle changes in new field environments during real field inference. Frontiers Media S.A. 2023-08-09 /pmc/articles/PMC10445656/ /pubmed/37621880 http://dx.doi.org/10.3389/fpls.2023.1234616 Text en Copyright © 2023 Ilyas, Lee, Won, Jeong and Kim 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
Ilyas, Talha
Lee, Jonghoon
Won, Okjae
Jeong, Yongchae
Kim, Hyongsuk
Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title_full Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title_fullStr Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title_full_unstemmed Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title_short Overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
title_sort overcoming field variability: unsupervised domain adaptation for enhanced crop-weed recognition in diverse farmlands
topic Plant Science
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10445656/
https://www.ncbi.nlm.nih.gov/pubmed/37621880
http://dx.doi.org/10.3389/fpls.2023.1234616
work_keys_str_mv AT ilyastalha overcomingfieldvariabilityunsuperviseddomainadaptationforenhancedcropweedrecognitionindiversefarmlands
AT leejonghoon overcomingfieldvariabilityunsuperviseddomainadaptationforenhancedcropweedrecognitionindiversefarmlands
AT wonokjae overcomingfieldvariabilityunsuperviseddomainadaptationforenhancedcropweedrecognitionindiversefarmlands
AT jeongyongchae overcomingfieldvariabilityunsuperviseddomainadaptationforenhancedcropweedrecognitionindiversefarmlands
AT kimhyongsuk overcomingfieldvariabilityunsuperviseddomainadaptationforenhancedcropweedrecognitionindiversefarmlands