Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images

Automatic crop interpretation can provide important reference information for national agricultural decision-making. However, due to the diverse characteristics and complex spatial relationship of crops, remote sensing images taken from a bird's eye view are insufficient in vertical featu...

Full description

Saved in:
Bibliographic Details
Main Authors: Wenyue Li, Bingfang Wu, Runyu Fan, Fuyou Tian, Miao Zhang, Zhaoying Zhou, Jun Hu, Ruyi Feng, Fangming Wu
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10731986/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850270287282569216
author Wenyue Li
Bingfang Wu
Runyu Fan
Fuyou Tian
Miao Zhang
Zhaoying Zhou
Jun Hu
Ruyi Feng
Fangming Wu
author_facet Wenyue Li
Bingfang Wu
Runyu Fan
Fuyou Tian
Miao Zhang
Zhaoying Zhou
Jun Hu
Ruyi Feng
Fangming Wu
author_sort Wenyue Li
collection DOAJ
description Automatic crop interpretation can provide important reference information for national agricultural decision-making. However, due to the diverse characteristics and complex spatial relationship of crops, remote sensing images taken from a bird's eye view are insufficient in vertical features of crops, making it difficult to interpret crop types and locations accurately. The similar features and blurred edges between different crops make it difficult to extract crop boundaries accurately. Due to the high memory and computational costs, most of the deep learning-based models face efficiency limitations in real-scenario crop interpretation. To address the abovementioned issues, this article proposes a novel lightweight neural network, namely the CropNet, for crop interpretation. Aiming at the problem of feature similarity among different categories of crops, this article designs a global-local path aggregation (GLPA) mechanism, which uses shallow and deep neural networks to extract global detail information and local high-level information to enhance feature discrimination. An edge context feature enhancement module (ECFEM) is proposed to enhance edge and context feature extraction to address the problem of ambiguous crop edges. Finally, a feature fusion module based on an attention mechanism is used to automatically weigh different feature channels to enhance the crop semantics. To demonstrate the effectiveness of the CropNet proposed in this article, we constructed several sets of comparison experiments comparing it with state-of-the-art deep learning models on a manually labeled vehicle-view crop image dataset. The experimental results show that CropNet has better semantic segmentation results with fewer model parameters and lower computational costs.
format Article
id doaj-art-c466cd6bb15146d49083553fdcbb9155
institution OA Journals
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-c466cd6bb15146d49083553fdcbb91552025-08-20T01:52:42ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-011849650910.1109/JSTARS.2024.348124810731986Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View ImagesWenyue Li0https://orcid.org/0009-0006-3516-1885Bingfang Wu1https://orcid.org/0000-0001-5546-365XRunyu Fan2https://orcid.org/0000-0002-5259-5670Fuyou Tian3https://orcid.org/0000-0003-1758-8763Miao Zhang4https://orcid.org/0000-0002-4021-2492Zhaoying Zhou5Jun Hu6Ruyi Feng7https://orcid.org/0000-0002-5709-690XFangming Wu8School of Computer Science, China University of Geosciences, Wuhan, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaSchool of Computer Science, China University of Geosciences, Wuhan, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaSchool of Computer Science, China University of Geosciences, Wuhan, ChinaSchool of Computer Science, China University of Geosciences, Wuhan, ChinaSchool of Computer Science, China University of Geosciences, Wuhan, ChinaKey Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAutomatic crop interpretation can provide important reference information for national agricultural decision-making. However, due to the diverse characteristics and complex spatial relationship of crops, remote sensing images taken from a bird's eye view are insufficient in vertical features of crops, making it difficult to interpret crop types and locations accurately. The similar features and blurred edges between different crops make it difficult to extract crop boundaries accurately. Due to the high memory and computational costs, most of the deep learning-based models face efficiency limitations in real-scenario crop interpretation. To address the abovementioned issues, this article proposes a novel lightweight neural network, namely the CropNet, for crop interpretation. Aiming at the problem of feature similarity among different categories of crops, this article designs a global-local path aggregation (GLPA) mechanism, which uses shallow and deep neural networks to extract global detail information and local high-level information to enhance feature discrimination. An edge context feature enhancement module (ECFEM) is proposed to enhance edge and context feature extraction to address the problem of ambiguous crop edges. Finally, a feature fusion module based on an attention mechanism is used to automatically weigh different feature channels to enhance the crop semantics. To demonstrate the effectiveness of the CropNet proposed in this article, we constructed several sets of comparison experiments comparing it with state-of-the-art deep learning models on a manually labeled vehicle-view crop image dataset. The experimental results show that CropNet has better semantic segmentation results with fewer model parameters and lower computational costs.https://ieeexplore.ieee.org/document/10731986/Deep learningmultifeature fusionsemantic segmentationcrop interpretation
spellingShingle Wenyue Li
Bingfang Wu
Runyu Fan
Fuyou Tian
Miao Zhang
Zhaoying Zhou
Jun Hu
Ruyi Feng
Fangming Wu
Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep learning
multifeature fusion
semantic segmentation
crop interpretation
title Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
title_full Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
title_fullStr Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
title_full_unstemmed Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
title_short Multiclass Crop Interpretation via a Lightweight Attentive Feature Fusion Network Using Vehicle-View Images
title_sort multiclass crop interpretation via a lightweight attentive feature fusion network using vehicle view images
topic Deep learning
multifeature fusion
semantic segmentation
crop interpretation
url https://ieeexplore.ieee.org/document/10731986/
work_keys_str_mv AT wenyueli multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT bingfangwu multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT runyufan multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT fuyoutian multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT miaozhang multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT zhaoyingzhou multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT junhu multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT ruyifeng multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages
AT fangmingwu multiclasscropinterpretationviaalightweightattentivefeaturefusionnetworkusingvehicleviewimages