Related Experiment Video
Updated: Sep 16, 2025
07:08
Identification of Novel Regulators of Plant Transpiration by Large-Scale Thermal Imaging Screening in Helianthus Annuus
Published on: January 30, 2020
6.1K
Intelligent Identification of Tea Plant Seedlings Under High-Temperature Conditions via YOLOv11-MEIP Model Based on
Chun Wang1,2, Zejun Wang2,3, Lijiao Chen2,3,4
1College of Mechanical and Electrical Engineering, Yunnan Agricultural University, Kunming 650201, China.
Plants (Basel, Switzerland)
|July 12, 2025
Summary
This study introduces an improved YOLOv11 model using chlorophyll fluorescence imaging to accurately identify high-temperature stress in tea plant seedlings. The new model enhances detection accuracy while reducing computational costs for efficient crop monitoring.
Area of Science:
- Agricultural Science
- Plant Physiology
- Computer Vision
- Machine Learning
Background:
- High-temperature stress significantly impacts tea plant seedling health and yield.
- Non-destructive, intelligent methods are needed for early stress detection in agriculture.
- Chlorophyll fluorescence imaging offers a sensitive, non-invasive approach to assess plant stress.
Purpose of the Study:
- To develop an efficient, non-destructive, and intelligent identification system for tea plant seedlings under high-temperature stress.
- To improve the YOLOv11 object detection model for enhanced accuracy and reduced computational load in stress level classification.
- To provide a technical reference for monitoring and preventing heat stress in tea gardens and other crops.
Main Methods:
- Acquisition of raw fluorescence images from tea plant seedlings subjected to varying high-temperature conditions.
- Utilized maximum photochemical efficiency (Fv/Fm) derived from fluorescence parameters to construct the dataset.
- Improved the YOLOv11 model by integrating a lightweight MobileNetV4 backbone, EUCB, iRMB, and PConv modules for efficient feature extraction and upsampling.
Main Results:
- The enhanced YOLOv11-MEIP model achieved superior performance with precision (99.25%), recall (99.19%), and mAP50 (99.46%).
- Significant improvements over the original YOLOv11 model: +4.05% precision, +7.86% recall, and +3.42% mAP50.
- Reduced model parameters by 29.45%, indicating enhanced computational efficiency.
Conclusions:
- The YOLOv11-MEIP model offers a novel intelligent method for classifying high-temperature stress levels in tea seedlings.
- The study demonstrates the effectiveness of integrating chlorophyll fluorescence imaging with improved deep learning models for crop stress monitoring.
- Provides valuable theoretical support and technical reference for managing heat stress in agricultural settings.
