Next Article in Journal
Detecting the Unseen: Hyperspectral Image Analysis for the Detection of Early Symptoms of Late Blight in Tomato Plants and Design of Its Machine Vision Application
Previous Article in Journal
Morphology-Informed Mechanical Design and Preliminary Evaluation of an Integrated Machine for Continuous Lettuce Postharvest Processing
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
AgriEngineering
Volume 8
Issue 9
10.3390/agriengineering8090353
agriengineering-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Article

CMAE-Unet: A Study on a U-Net-Based Model for Semantic Segmentation of Unripe Tomato Images

by
Jianhua Zheng
1,2,3,
Huanghui Zhao
4,
Xiaoshan Ma
1,
Guiming Huang
1,
Yongshen Liang
1,
Jinfang Liu
1,
Zhaoxi Luo
1,
Yuanlan Ye
1,* and
Jianru Chen
1,*
1
College of Artificial Intelligence, Zhongkai University of Agriculture and Engineering, Guangzhou 510225, China
2
Guangzhou Key Laboratory of Agricultural Products Quality & Safety Traceability Information Technology, Zhongkai University of Agriculture and Engineering, Guangzhou 510225, China
3
Smart Agriculture Innovation Research Institute, Zhongkai University of Agriculture and Engineering, Guangzhou 510225, China
4
College of Artificial Intelligence, South China Agricultural University, Guangzhou 510642, China
*
Authors to whom correspondence should be addressed.
AgriEngineering 2026, 8(9), 353; https://doi.org/10.3390/agriengineering8090353
Submission received: 6 July 2026 / Revised: 1 August 2026 / Accepted: 21 August 2026 / Published: 25 August 2026
(This article belongs to the Section Computer Applications and Artificial Intelligence in Agriculture)
Versions Notes

Abstract

In natural scenes, the high resemblance between unripe tomatoes and foliage, combined with severe occlusion, challenges current semantic segmentation models, causing poor accuracy and indistinct boundaries. To overcome this, we propose CMAE-UNet, a camouflage suppression and edge enhancement model based on the UNet architecture. Specifically, a Global-Local Integrated Spatial Attention (GLISA) encoder merges dual-branch dilated convolutions, residual structures, and an Efficient Multi-scale Attention mechanism to expand receptive fields and highlight targets in complex backgrounds. Furthermore, a Frequency-Domain Feature Enhancement (FFE) module leverages the Fast Fourier Transform to separate and adaptively enhance distinct frequency components, effectively mitigating camouflage interference. Additionally, a Directional Edge Enhancement (DEE) module uses three-directional learnable convolutions and spatial attention to sharpen indistinct target contours. Evaluated on a custom Tomato dataset encompassing five complex scenarios, CMAE-UNet outperforms 12 prominent methods in mIoU, Dice, and Sen metrics, yielding smoother and more precise segmentation boundaries. The model robustly withstands field interference, providing strong technological support for automated tomato detection, intelligent harvesting, and growth monitoring.
Keywords: precision agriculture; deep learning; camouflage suppression; contour refinement precision agriculture; deep learning; camouflage suppression; contour refinement

Share and Cite

MDPI and ACS Style

Zheng, J.; Zhao, H.; Ma, X.; Huang, G.; Liang, Y.; Liu, J.; Luo, Z.; Ye, Y.; Chen, J. CMAE-Unet: A Study on a U-Net-Based Model for Semantic Segmentation of Unripe Tomato Images. AgriEngineering 2026, 8, 353. https://doi.org/10.3390/agriengineering8090353

AMA Style

Zheng J, Zhao H, Ma X, Huang G, Liang Y, Liu J, Luo Z, Ye Y, Chen J. CMAE-Unet: A Study on a U-Net-Based Model for Semantic Segmentation of Unripe Tomato Images. AgriEngineering. 2026; 8(9):353. https://doi.org/10.3390/agriengineering8090353

Chicago/Turabian Style

Zheng, Jianhua, Huanghui Zhao, Xiaoshan Ma, Guiming Huang, Yongshen Liang, Jinfang Liu, Zhaoxi Luo, Yuanlan Ye, and Jianru Chen. 2026. "CMAE-Unet: A Study on a U-Net-Based Model for Semantic Segmentation of Unripe Tomato Images" AgriEngineering 8, no. 9: 353. https://doi.org/10.3390/agriengineering8090353

APA Style

Zheng, J., Zhao, H., Ma, X., Huang, G., Liang, Y., Liu, J., Luo, Z., Ye, Y., & Chen, J. (2026). CMAE-Unet: A Study on a U-Net-Based Model for Semantic Segmentation of Unripe Tomato Images. AgriEngineering, 8(9), 353. https://doi.org/10.3390/agriengineering8090353

Article Metrics

Back to TopTop