Cargando…

Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images

Existing pollen identification methods heavily rely on the scale and quality of pollen images. However, there are many impurities in real-world SEM images that should be considered. This paper proposes a collaborative learning method to jointly improve the performance of pollen segmentation and clas...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Jianqiang, Xu, Qinlan, Cheng, Wenxiu, Zhao, Linna, Liu, Suqin, Gao, Zhengkai, Xu, Xi, Ye, Caihua, You, Huanling
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9867018/
https://www.ncbi.nlm.nih.gov/pubmed/36676197
http://dx.doi.org/10.3390/life13010247
_version_ 1784876237862207488
author Li, Jianqiang
Xu, Qinlan
Cheng, Wenxiu
Zhao, Linna
Liu, Suqin
Gao, Zhengkai
Xu, Xi
Ye, Caihua
You, Huanling
author_facet Li, Jianqiang
Xu, Qinlan
Cheng, Wenxiu
Zhao, Linna
Liu, Suqin
Gao, Zhengkai
Xu, Xi
Ye, Caihua
You, Huanling
author_sort Li, Jianqiang
collection PubMed
description Existing pollen identification methods heavily rely on the scale and quality of pollen images. However, there are many impurities in real-world SEM images that should be considered. This paper proposes a collaborative learning method to jointly improve the performance of pollen segmentation and classification in a weakly supervised manner. It first locates pollen regions from the raw images based on the detection model. To improve the classification performance, we segmented the pollen grains through a pre-trained U-Net using unsupervised pollen contour features. The segmented pollen regions were fed into a deep convolutional neural network to obtain the activation maps, which were used to further refine the segmentation masks. In this way, both segmentation and classification models can be collaboratively trained, supervised by just pollen contour features and class-specific information. Extensive experiments on real-world datasets were conducted, and the results prove that our method effectively avoids impurity interference and improves pollen identification accuracy (86.6%) under the limited supervision (around 1000 images with image-level labels).
format Online
Article
Text
id pubmed-9867018
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-98670182023-01-22 Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images Li, Jianqiang Xu, Qinlan Cheng, Wenxiu Zhao, Linna Liu, Suqin Gao, Zhengkai Xu, Xi Ye, Caihua You, Huanling Life (Basel) Article Existing pollen identification methods heavily rely on the scale and quality of pollen images. However, there are many impurities in real-world SEM images that should be considered. This paper proposes a collaborative learning method to jointly improve the performance of pollen segmentation and classification in a weakly supervised manner. It first locates pollen regions from the raw images based on the detection model. To improve the classification performance, we segmented the pollen grains through a pre-trained U-Net using unsupervised pollen contour features. The segmented pollen regions were fed into a deep convolutional neural network to obtain the activation maps, which were used to further refine the segmentation masks. In this way, both segmentation and classification models can be collaboratively trained, supervised by just pollen contour features and class-specific information. Extensive experiments on real-world datasets were conducted, and the results prove that our method effectively avoids impurity interference and improves pollen identification accuracy (86.6%) under the limited supervision (around 1000 images with image-level labels). MDPI 2023-01-16 /pmc/articles/PMC9867018/ /pubmed/36676197 http://dx.doi.org/10.3390/life13010247 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
Li, Jianqiang
Xu, Qinlan
Cheng, Wenxiu
Zhao, Linna
Liu, Suqin
Gao, Zhengkai
Xu, Xi
Ye, Caihua
You, Huanling
Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title_full Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title_fullStr Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title_full_unstemmed Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title_short Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images
title_sort weakly supervised collaborative learning for airborne pollen segmentation and classification from sem images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9867018/
https://www.ncbi.nlm.nih.gov/pubmed/36676197
http://dx.doi.org/10.3390/life13010247
work_keys_str_mv AT lijianqiang weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT xuqinlan weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT chengwenxiu weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT zhaolinna weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT liusuqin weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT gaozhengkai weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT xuxi weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT yecaihua weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages
AT youhuanling weaklysupervisedcollaborativelearningforairbornepollensegmentationandclassificationfromsemimages