Related Experiment Video
Updated: Nov 23, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
1.2K
Detection of Strawberry Diseases Using a Convolutional Neural Network
Jia-Rong Xiao1, Pei-Che Chung2, Hung-Yi Wu3
1Department of Mechanical Engineering, National United University, Miaoli 360001, Taiwan.
Plants (Basel, Switzerland)
|December 30, 2020
Summary
This study introduces a convolutional neural network (CNN) model for detecting strawberry diseases like leaf blight, gray mold, and powdery mildew. The AI model achieved high accuracy, offering a cost-effective solution for disease detection in agriculture.
Area of Science:
- Agricultural Science
- Computer Science
- Plant Pathology
Background:
- Strawberry cultivation in Taiwan, particularly in Miaoli County, faces significant economic losses due to various diseases.
- Anthracnose crown rot and other leaf and fruit diseases have caused substantial damage to strawberry seedlings and plants.
- Automated agriculture and image recognition are crucial for efficient and early detection of these diseases.
Purpose of the Study:
- To develop an automated image recognition technique for detecting key strawberry diseases.
- To utilize a convolutional neural network (CNN) model for enhanced disease identification.
- To provide a reliable and cost-effective method for strawberry disease management.
Main Methods:
- Development of a convolutional neural network (CNN) model for image recognition.
- Utilizing two datasets: original images and feature images for training the model.
- Employing the ResNet50 architecture with a training period of 20 epochs.
Main Results:
- The CNN model achieved 100% accuracy for detecting leaf blight (crown, leaf, fruit), 98% for gray mold, and 98% for powdery mildew.
- The model trained on feature images achieved 99.60% accuracy, significantly outperforming the 1.53% accuracy from original images.
- The ResNet50 model demonstrated high efficacy in classifying strawberry diseases from images.
Conclusions:
- The proposed CNN-based image recognition technique is a simple, reliable, and cost-effective tool for strawberry disease detection.
- Automated disease detection using AI can significantly aid in managing crop health and reducing economic losses.
- This technology has the potential to improve strawberry farming practices and sustainability.
