Related Experiment Video
Updated: Oct 5, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
938
Recognition of Leaf Disease Using Hybrid Convolutional Neural Network by Applying Feature Reduction
Prabhjot Kaur1, Shilpi Harnal1, Rajeev Tiwari2
1Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura 140401, Punjab, India.
Sensors (Basel, Switzerland)
|January 22, 2022
Summary
This study introduces an automated method for detecting four grapevine diseases using EfficientNet B7 deep learning and logistic regression. The approach achieves 98.7% accuracy, improving plant disease diagnosis for sustainable agriculture.
Area of Science:
- Agricultural Science
- Computer Vision
- Machine Learning
Background:
- Plant diseases pose significant threats to food security and agricultural economies, necessitating accurate and efficient detection methods.
- Manual monitoring of grapevine diseases is labor-intensive and often insufficient due to disease diversity and time constraints.
- Automated plant disease characterization is crucial for sustainable agricultural practices and input management.
Purpose of the Study:
- To develop and evaluate an automated system for the comprehensive detection of four major grapevine diseases: Leaf blight, Black rot, Stable, and Black measles.
- To address the limitation of previous studies that focused on detecting only one or two diseases.
- To investigate the efficacy of deep learning and machine learning techniques for accurate plant disease identification.
Main Methods:
- Utilized the Plant Village dataset for training and validation.
- Employed transfer learning to retrain the EfficientNet B7 deep architecture.
- Applied Logistic Regression for feature down-sampling, followed by state-of-the-art classifiers for final disease identification.
Main Results:
- The proposed method achieved a high accuracy of 98.7% in detecting the four specified grapevine diseases after 92 epochs.
- The study successfully identified the most discriminant features for accurate disease classification.
- A comparative analysis confirmed the effectiveness of the proposed technique against existing methods.
Conclusions:
- The developed automated system offers a robust and accurate solution for detecting multiple grapevine diseases, surpassing previous research limitations.
- The combination of EfficientNet B7 and Logistic Regression provides a powerful approach for image-based plant disease diagnosis.
- The findings suggest a suitable classifier for practical application, contributing to sustainable agriculture through improved disease management.
