Cargando…
A novel method for maize leaf disease classification using the RGB-D post-segmentation image data
Maize (Zea mays L.) is one of the most important crops, influencing food production and even the whole industry. In recent years, global crop production has been facing great challenges from diseases. However, most of the traditional methods make it difficult to efficiently identify disease-related...
Autores principales: | , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10562608/ https://www.ncbi.nlm.nih.gov/pubmed/37822341 http://dx.doi.org/10.3389/fpls.2023.1268015 |
_version_ | 1785118165930344448 |
---|---|
author | Nan, Fei Song, Yang Yu, Xun Nie, Chenwei Liu, Yadong Bai, Yali Zou, Dongxiao Wang, Chao Yin, Dameng Yang, Wude Jin, Xiuliang |
author_facet | Nan, Fei Song, Yang Yu, Xun Nie, Chenwei Liu, Yadong Bai, Yali Zou, Dongxiao Wang, Chao Yin, Dameng Yang, Wude Jin, Xiuliang |
author_sort | Nan, Fei |
collection | PubMed |
description | Maize (Zea mays L.) is one of the most important crops, influencing food production and even the whole industry. In recent years, global crop production has been facing great challenges from diseases. However, most of the traditional methods make it difficult to efficiently identify disease-related phenotypes in germplasm resources, especially in actual field environments. To overcome this limitation, our study aims to evaluate the potential of the multi-sensor synchronized RGB-D camera with depth information for maize leaf disease classification. We distinguished maize leaves from the background based on the RGB-D depth information to eliminate interference from complex field environments. Four deep learning models (i.e., Resnet50, MobilenetV2, Vgg16, and Efficientnet-B3) were used to classify three main types of maize diseases, i.e., the curvularia leaf spot [Curvularia lunata (Wakker) Boedijn], the small spot [Bipolaris maydis (Nishik.) Shoemaker], and the mixed spot diseases. We finally compared the pre-segmentation and post-segmentation results to test the robustness of the above models. Our main findings are: 1) The maize disease classification models based on the pre-segmentation image data performed slightly better than the ones based on the post-segmentation image data. 2) The pre-segmentation models overestimated the accuracy of disease classification due to the complexity of the background, but post-segmentation models focusing on leaf disease features provided more practical results with shorter prediction times. 3) Among the post-segmentation models, the Resnet50 and MobilenetV2 models showed similar accuracy and were better than the Vgg16 and Efficientnet-B3 models, and the MobilenetV2 model performed better than the other three models in terms of the size and the single image prediction time. Overall, this study provides a novel method for maize leaf disease classification using the post-segmentation image data from a multi-sensor synchronized RGB-D camera and offers the possibility of developing relevant portable devices. |
format | Online Article Text |
id | pubmed-10562608 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-105626082023-10-11 A novel method for maize leaf disease classification using the RGB-D post-segmentation image data Nan, Fei Song, Yang Yu, Xun Nie, Chenwei Liu, Yadong Bai, Yali Zou, Dongxiao Wang, Chao Yin, Dameng Yang, Wude Jin, Xiuliang Front Plant Sci Plant Science Maize (Zea mays L.) is one of the most important crops, influencing food production and even the whole industry. In recent years, global crop production has been facing great challenges from diseases. However, most of the traditional methods make it difficult to efficiently identify disease-related phenotypes in germplasm resources, especially in actual field environments. To overcome this limitation, our study aims to evaluate the potential of the multi-sensor synchronized RGB-D camera with depth information for maize leaf disease classification. We distinguished maize leaves from the background based on the RGB-D depth information to eliminate interference from complex field environments. Four deep learning models (i.e., Resnet50, MobilenetV2, Vgg16, and Efficientnet-B3) were used to classify three main types of maize diseases, i.e., the curvularia leaf spot [Curvularia lunata (Wakker) Boedijn], the small spot [Bipolaris maydis (Nishik.) Shoemaker], and the mixed spot diseases. We finally compared the pre-segmentation and post-segmentation results to test the robustness of the above models. Our main findings are: 1) The maize disease classification models based on the pre-segmentation image data performed slightly better than the ones based on the post-segmentation image data. 2) The pre-segmentation models overestimated the accuracy of disease classification due to the complexity of the background, but post-segmentation models focusing on leaf disease features provided more practical results with shorter prediction times. 3) Among the post-segmentation models, the Resnet50 and MobilenetV2 models showed similar accuracy and were better than the Vgg16 and Efficientnet-B3 models, and the MobilenetV2 model performed better than the other three models in terms of the size and the single image prediction time. Overall, this study provides a novel method for maize leaf disease classification using the post-segmentation image data from a multi-sensor synchronized RGB-D camera and offers the possibility of developing relevant portable devices. Frontiers Media S.A. 2023-09-26 /pmc/articles/PMC10562608/ /pubmed/37822341 http://dx.doi.org/10.3389/fpls.2023.1268015 Text en Copyright © 2023 Nan, Song, Yu, Nie, Liu, Bai, Zou, Wang, Yin, Yang and Jin 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 Nan, Fei Song, Yang Yu, Xun Nie, Chenwei Liu, Yadong Bai, Yali Zou, Dongxiao Wang, Chao Yin, Dameng Yang, Wude Jin, Xiuliang A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title | A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title_full | A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title_fullStr | A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title_full_unstemmed | A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title_short | A novel method for maize leaf disease classification using the RGB-D post-segmentation image data |
title_sort | novel method for maize leaf disease classification using the rgb-d post-segmentation image data |
topic | Plant Science |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10562608/ https://www.ncbi.nlm.nih.gov/pubmed/37822341 http://dx.doi.org/10.3389/fpls.2023.1268015 |
work_keys_str_mv | AT nanfei anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT songyang anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yuxun anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT niechenwei anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT liuyadong anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT baiyali anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT zoudongxiao anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT wangchao anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yindameng anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yangwude anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT jinxiuliang anovelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT nanfei novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT songyang novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yuxun novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT niechenwei novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT liuyadong novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT baiyali novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT zoudongxiao novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT wangchao novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yindameng novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT yangwude novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata AT jinxiuliang novelmethodformaizeleafdiseaseclassificationusingthergbdpostsegmentationimagedata |