Related Experiment Video
Updated: May 10, 2025
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
1.2K
Intelligent Recognition of Goji Berry Pests Using CNN With Multi-Graphic-Occlusion Data Augmentation and Multiple
1School of Electronics and Information Engineering, Hangzhou Dianzi University, Hangzhou, China.
Archives of Insect Biochemistry and Physiology
|April 22, 2025
Summary
A new deep learning model, GojiNet, accurately identifies 17 goji berry pests with 95.35% accuracy. This AI approach improves upon traditional methods, offering a faster and more precise solution for crop protection.
Area of Science:
- Agricultural Science
- Computer Science
- Artificial Intelligence
Background:
- Pest infestations significantly threaten goji berry yield and quality.
- Manual pest identification is subjective, time-consuming, and labor-intensive.
Purpose of the Study:
- To develop an improved convolutional neural network (CNN) for accurate identification of 17 goji berry pest types.
- To enhance pest identification efficiency and accuracy in goji berry cultivation.
Main Methods:
- Data augmentation using a multi-graph-occlusion technique.
- Development of GojiNet, a novel CNN based on ResNet18 with multi-attention fusion modules.
- Training and evaluation of the GojiNet model on a dataset of goji berry pests.
Main Results:
- GojiNet achieved an average recognition accuracy of 95.35%.
- This represents a 2.60% improvement over the baseline ResNet18 network.
- The model demonstrated enhanced accuracy with only a slight increase in training time and reduced model size.
Conclusions:
- GojiNet offers a highly accurate and efficient solution for goji berry pest identification.
- Deep learning holds significant potential for intelligent and precise pest management in agriculture.
- The study provides a referential solution for automated pest identification systems.
