CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading

Tobacco leaf grading plays a crucial role in ensuring the quality of tobacco production. For a very long period, the grading process is manually determined by experienced experts. In recent years, some methods have been introduced to automate the grading process by utilizing the reflection images of...

Full description

Saved in:
Bibliographic Details
Main Authors: Qinglin He, Xiaobing Zhang, Jianxin Hu, Zehua Sheng, Qi Li, Si-Yuan Cao, Hui-Liang Shen
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10268436/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850144822400122880
author Qinglin He
Xiaobing Zhang
Jianxin Hu
Zehua Sheng
Qi Li
Si-Yuan Cao
Hui-Liang Shen
author_facet Qinglin He
Xiaobing Zhang
Jianxin Hu
Zehua Sheng
Qi Li
Si-Yuan Cao
Hui-Liang Shen
author_sort Qinglin He
collection DOAJ
description Tobacco leaf grading plays a crucial role in ensuring the quality of tobacco production. For a very long period, the grading process is manually determined by experienced experts. In recent years, some methods have been introduced to automate the grading process by utilizing the reflection images of tobacco leaves. However, the high visual similarity among reflection images at different grades renders a single reflection image insufficient for achieving accurate grading. Besides, the tobacco leaves with an identical grade may have inconsistent visual appearances due to their different planting locations. It is known that an expert integrates multimodal information such as visual, tactile, and planting location cues when performing grading. Inspired by this, we propose an end-to-end Cross-modal Enhancement Network, named CMENet, for automatic tobacco leaf grading. In addition to the common reflection image, the network also adopts a transmission image to incorporate the thickness information in manual grading. CMENet comprises a difference-aware fusion module and a meta self-attention module, enabling the extraction of multimodal information from the transmission image and the planting location, respectively. Experimental results demonstrate that our CMENet achieves a high grading accuracy (80.15%) when incorporating multimodal information, surpassing the performance of existing methods that rely solely on reflection images.
format Article
id doaj-art-ae1e6cdfe2624aabb9a0cfeaad06d479
institution OA Journals
issn 2169-3536
language English
publishDate 2023-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-ae1e6cdfe2624aabb9a0cfeaad06d4792025-08-20T02:28:15ZengIEEEIEEE Access2169-35362023-01-011110920110921210.1109/ACCESS.2023.332111110268436CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf GradingQinglin He0Xiaobing Zhang1Jianxin Hu2Zehua Sheng3https://orcid.org/0000-0002-1721-9143Qi Li4Si-Yuan Cao5https://orcid.org/0000-0001-5143-4501Hui-Liang Shen6https://orcid.org/0000-0001-8469-019XCollege of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, ChinaChina Tobacco Zhejiang Industrial Company Ltd., Hangzhou, ChinaCollege of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, ChinaCollege of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, ChinaChina Tobacco Zhejiang Industrial Company Ltd., Hangzhou, ChinaCollege of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, ChinaCollege of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, ChinaTobacco leaf grading plays a crucial role in ensuring the quality of tobacco production. For a very long period, the grading process is manually determined by experienced experts. In recent years, some methods have been introduced to automate the grading process by utilizing the reflection images of tobacco leaves. However, the high visual similarity among reflection images at different grades renders a single reflection image insufficient for achieving accurate grading. Besides, the tobacco leaves with an identical grade may have inconsistent visual appearances due to their different planting locations. It is known that an expert integrates multimodal information such as visual, tactile, and planting location cues when performing grading. Inspired by this, we propose an end-to-end Cross-modal Enhancement Network, named CMENet, for automatic tobacco leaf grading. In addition to the common reflection image, the network also adopts a transmission image to incorporate the thickness information in manual grading. CMENet comprises a difference-aware fusion module and a meta self-attention module, enabling the extraction of multimodal information from the transmission image and the planting location, respectively. Experimental results demonstrate that our CMENet achieves a high grading accuracy (80.15%) when incorporating multimodal information, surpassing the performance of existing methods that rely solely on reflection images.https://ieeexplore.ieee.org/document/10268436/Tobacco leaf gradingimage classificationconvolutional neural networkcross-modal information fusion
spellingShingle Qinglin He
Xiaobing Zhang
Jianxin Hu
Zehua Sheng
Qi Li
Si-Yuan Cao
Hui-Liang Shen
CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
IEEE Access
Tobacco leaf grading
image classification
convolutional neural network
cross-modal information fusion
title CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
title_full CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
title_fullStr CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
title_full_unstemmed CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
title_short CMENet: A Cross-Modal Enhancement Network for Tobacco Leaf Grading
title_sort cmenet a cross modal enhancement network for tobacco leaf grading
topic Tobacco leaf grading
image classification
convolutional neural network
cross-modal information fusion
url https://ieeexplore.ieee.org/document/10268436/
work_keys_str_mv AT qinglinhe cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT xiaobingzhang cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT jianxinhu cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT zehuasheng cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT qili cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT siyuancao cmenetacrossmodalenhancementnetworkfortobaccoleafgrading
AT huiliangshen cmenetacrossmodalenhancementnetworkfortobaccoleafgrading