Related Experiment Video
Updated: Jul 10, 2025
08:21
Real-time In Vivo Recording of Arabidopsis Calcium Signals During Insect Feeding Using a Fluorescent Biosensor
Published on: August 15, 2017
12.9K
Aphid Recognition and Counting Based on an Improved YOLOv5 Algorithm in a Climate Chamber Environment.
Xiaoyin Li1, Lixing Wang1, Hong Miao1
1College of Mechanical Engineering, Yangzhou University, Yangzhou 225127, China.
Insects
|November 24, 2023
Summary
This study introduces an enhanced YOLOv5 model for accurate aphid counting and recognition in climate chambers. The improved model achieves high accuracy and speed, offering a robust solution for pest control.
Area of Science:
- Agricultural Science
- Computer Vision
- Machine Learning
Background:
- Accurate aphid identification and counting in climate chambers are challenging due to variable light intensity, aphid aggregation, and small target sizes.
- Existing methods struggle with the complexities of controlled environment agriculture (CEA) pest monitoring.
Purpose of the Study:
- To develop an improved YOLOv5 model for accurate aphid recognition and counting in climate chambers.
- To enhance the robustness and efficiency of automated pest detection systems for CEA.
Main Methods:
- An improved YOLOv5 model incorporating Mosaic and GridMask for data augmentation to mitigate overfitting.
- Integration of Convolutional Block Attention Module (CBAM) in the backbone for enhanced small target recognition.
- Utilized Bi-directional Feature Pyramid Network (BiFPN) for feature fusion in the YOLOv5 neck.
- Introduced a Transformer structure before the detection head to analyze the impact of environmental factors on recognition.
Main Results:
- The proposed model achieved a recognition accuracy and recall rate of 99.1% and a mean Average Precision (mAP)@0.5 of 99.3%.
- Inference time was reduced to 9.4 ms, outperforming other YOLO series networks.
- Demonstrated strong robustness in real-world recognition tasks within climate chambers.
Conclusions:
- The enhanced YOLOv5 model significantly improves aphid detection accuracy and efficiency in challenging climate chamber environments.
- This model provides a valuable tool for automated pest prevention and control strategies in agricultural research and practice.
