Related Concept Videos
Survival Tree
Building a Survival Tree
Constructing a...
Light Acquisition
Responses to Gravity and Touch
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
A lightweight explainable deep learning framework for coal classification.
Optimized multi-modal action recognition using multi-agent systems and adaptive temporal attention.
A framework based on hybrid Suzuki-Abe and convex Hull approach for improved classification of skin lesions.
Related Experiment Video
Updated: Jun 28, 2025

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
Deep transfer learning with gravitational search algorithm for enhanced plant disease classification.
Mehdhar S A M Al-Gaashani1, Nagwan Abdel Samee2, Reem Alkanhel2
1School of Resources and Environment, University of Electronic Science and Technology of China, 4 1st Ring Rd East 2 Section, Chenghua District, Chengdu, 610056, Sichuan, China.
This study introduces a new method for early plant disease identification using transfer learning and Gravitational Search Algorithm (GSA) optimization, achieving 99.2% precision. The approach significantly reduces features, enhancing efficiency for global food security.
Area of Science:
- Agricultural Science
- Computer Science
- Data Science
Background:
- Plant diseases pose a significant threat to global food security, causing substantial crop damage and economic losses.
- Early and accurate identification of plant diseases is crucial for effective management and mitigation strategies.
Purpose of the Study:
- To develop and evaluate a novel computational method for the early identification and classification of plant diseases.
- To enhance the efficiency and accuracy of plant disease diagnosis through advanced feature extraction and optimization techniques.
Main Methods:
- Utilized transfer learning with pretrained models (MobileNetV2, ResNe50V2) for multilayer feature extraction from plant leaf images.
- Employed the Gravitational Search Algorithm (GSA) for optimizing extracted features, followed by Multinomial Logistic Regression (MLR) for classification.
- Compared GSA optimization with Genetic Algorithm (GA) and contrasted MLR with K-Nearest Neighbors (KNN) for performance evaluation.
Main Results:
- The proposed GSA-optimized model achieved high classification precision, with an average of 99.2% for MLR and 98.6% for KNN.
- Demonstrated a significant reduction in feature count by over 50% without compromising diagnostic accuracy.
- Outperformed models using GA-optimized features, confirming the superiority of the GSA-based approach.
Conclusions:
- The developed method offers a robust and efficient solution for early plant disease detection, integrating sophisticated computational techniques into agriculture.
- This data-driven approach enhances plant health management strategies, contributing to improved worldwide food security.
- The significant feature reduction highlights the method's efficiency, reducing processing demands for practical agricultural applications.

