Cargando…

Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images

Accurate segmentation and detection of rice seedlings is essential for precision agriculture and high-yield cultivation. However, current methods suffer from high computational complexity and poor robustness to different rice varieties and densities. This article proposes 2 lightweight neural networ...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Panli, Sun, Xiaobo, Zhang, Donghui, Yang, Yuechao, Wang, Zhenhua
Formato: Online Artículo Texto
Lenguaje:English
Publicado: AAAS 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10688663/
https://www.ncbi.nlm.nih.gov/pubmed/38047001
http://dx.doi.org/10.34133/plantphenomics.0123
_version_ 1785152209765269504
author Zhang, Panli
Sun, Xiaobo
Zhang, Donghui
Yang, Yuechao
Wang, Zhenhua
author_facet Zhang, Panli
Sun, Xiaobo
Zhang, Donghui
Yang, Yuechao
Wang, Zhenhua
author_sort Zhang, Panli
collection PubMed
description Accurate segmentation and detection of rice seedlings is essential for precision agriculture and high-yield cultivation. However, current methods suffer from high computational complexity and poor robustness to different rice varieties and densities. This article proposes 2 lightweight neural network architectures, LW-Segnet and LW-Unet, for high-precision rice seedling segmentation. The networks adopt an encoder–decoder structure with hybrid lightweight convolutions and spatial pyramid dilated convolutions, achieving accurate segmentation while reducing model parameters. Multispectral imagery acquired by unmanned aerial vehicle (UAV) was used to train and test the models covering 3 rice varieties and different planting densities. Experimental results demonstrate that the proposed LW-Segnet and LW-Unet models achieve higher F1-scores and intersection over union values for seedling detection and row segmentation across varieties, indicating improved segmentation accuracy. Furthermore, the models exhibit stable performance when handling different varieties and densities, showing strong robustness. In terms of efficiency, the networks have lower graphics processing unit memory usage, complexity, and parameters but faster inference speeds, reflecting higher computational efficiency. In particular, the fast speed of LW-Unet indicates potential for real-time applications. The study presents lightweight yet effective neural network architectures for agricultural tasks. By handling multiple rice varieties and densities with high accuracy, efficiency, and robustness, the models show promise for use in edge devices and UAVs to assist precision farming and crop management. The findings provide valuable insights into designing lightweight deep learning models to tackle complex agricultural problems.
format Online
Article
Text
id pubmed-10688663
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher AAAS
record_format MEDLINE/PubMed
spelling pubmed-106886632023-12-01 Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images Zhang, Panli Sun, Xiaobo Zhang, Donghui Yang, Yuechao Wang, Zhenhua Plant Phenomics Research Article Accurate segmentation and detection of rice seedlings is essential for precision agriculture and high-yield cultivation. However, current methods suffer from high computational complexity and poor robustness to different rice varieties and densities. This article proposes 2 lightweight neural network architectures, LW-Segnet and LW-Unet, for high-precision rice seedling segmentation. The networks adopt an encoder–decoder structure with hybrid lightweight convolutions and spatial pyramid dilated convolutions, achieving accurate segmentation while reducing model parameters. Multispectral imagery acquired by unmanned aerial vehicle (UAV) was used to train and test the models covering 3 rice varieties and different planting densities. Experimental results demonstrate that the proposed LW-Segnet and LW-Unet models achieve higher F1-scores and intersection over union values for seedling detection and row segmentation across varieties, indicating improved segmentation accuracy. Furthermore, the models exhibit stable performance when handling different varieties and densities, showing strong robustness. In terms of efficiency, the networks have lower graphics processing unit memory usage, complexity, and parameters but faster inference speeds, reflecting higher computational efficiency. In particular, the fast speed of LW-Unet indicates potential for real-time applications. The study presents lightweight yet effective neural network architectures for agricultural tasks. By handling multiple rice varieties and densities with high accuracy, efficiency, and robustness, the models show promise for use in edge devices and UAVs to assist precision farming and crop management. The findings provide valuable insights into designing lightweight deep learning models to tackle complex agricultural problems. AAAS 2023-11-30 /pmc/articles/PMC10688663/ /pubmed/38047001 http://dx.doi.org/10.34133/plantphenomics.0123 Text en Copyright © 2023 Panli Zhang et al. https://creativecommons.org/licenses/by/4.0/Exclusive licensee Nanjing Agricultural University. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution License 4.0 (CC BY 4.0) (https://creativecommons.org/licenses/by/4.0/) .
spellingShingle Research Article
Zhang, Panli
Sun, Xiaobo
Zhang, Donghui
Yang, Yuechao
Wang, Zhenhua
Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title_full Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title_fullStr Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title_full_unstemmed Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title_short Lightweight Deep Learning Models for High-Precision Rice Seedling Segmentation from UAV-Based Multispectral Images
title_sort lightweight deep learning models for high-precision rice seedling segmentation from uav-based multispectral images
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10688663/
https://www.ncbi.nlm.nih.gov/pubmed/38047001
http://dx.doi.org/10.34133/plantphenomics.0123
work_keys_str_mv AT zhangpanli lightweightdeeplearningmodelsforhighprecisionriceseedlingsegmentationfromuavbasedmultispectralimages
AT sunxiaobo lightweightdeeplearningmodelsforhighprecisionriceseedlingsegmentationfromuavbasedmultispectralimages
AT zhangdonghui lightweightdeeplearningmodelsforhighprecisionriceseedlingsegmentationfromuavbasedmultispectralimages
AT yangyuechao lightweightdeeplearningmodelsforhighprecisionriceseedlingsegmentationfromuavbasedmultispectralimages
AT wangzhenhua lightweightdeeplearningmodelsforhighprecisionriceseedlingsegmentationfromuavbasedmultispectralimages