Related Experiment Video
Updated: Jul 2, 2025
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
1.4K
Using transfer learning-based plant disease classification and detection for sustainable agriculture.
Wasswa Shafik1, Ali Tufail2, Chandratilak De Silva Liyanage1
1School of Digital Science, Universiti Brunei Darussalam, Tungku Link, Gadong, BE1410, Brunei.
BMC Plant Biology
|February 26, 2024
Summary
This study introduces two advanced plant disease detection models (PDDNet-AE and PDDNet-LVE) that significantly improve early identification of crop diseases using deep transfer learning. These models achieve high accuracy, aiding global food security and sustainable agriculture.
Area of Science:
- Agricultural Science
- Computer Vision
- Machine Learning
Background:
- Accurate and timely plant disease detection is crucial for global food security and sustainable agriculture, aligning with UN Sustainable Development Goals.
- Existing methods for disease detection face challenges like overfitting and difficulties in fine feature extraction.
- Manual laboratory diagnosis is costly, time-consuming, and labor-intensive, highlighting the need for automated solutions in smart agriculture.
Purpose of the Study:
- To develop and evaluate novel deep learning models for efficient and accurate early plant disease detection and classification.
- To address the limitations of current methods by leveraging deep transfer learning and ensemble techniques for improved recognition accuracy.
Main Methods:
- Introduced two plant disease detection (PDDNet) models: early fusion (AE) and lead voting ensemble (LVE).
- Integrated nine pre-trained convolutional neural networks (CNNs) including DenseNet201, ResNet101, and EfficientNetB7, fine-tuned by deep feature extraction.
- Utilized the PlantVillage dataset (54,305 images, 15 classes) and a logistic regression classifier for performance evaluation and comparative analysis.
Main Results:
- The proposed PDDNet-AE and PDDNet-LVE models achieved high detection accuracies of 96.74% and 97.79%, respectively.
- These models demonstrated superior robustness and generalization capabilities compared to individual CNNs and state-of-the-art methods.
- The ensemble approach effectively mitigated complexities in fine feature extraction and overfitting issues.
Conclusions:
- The PDDNet models offer a significant advancement in automated plant disease detection, enhancing efficiency and accuracy.
- This research provides a robust framework for smart, sustainable agriculture, contributing to improved crop management and food security.
- The findings validate the effectiveness of deep transfer learning and ensemble methods for complex image classification tasks in agriculture.
Related Concept Videos
01:59Plant Breeding and Biotechnology
18.9K
Crop cultivation has a long history in human civilization, with records showing the cultivation of cereal plants beginning at around 8000 BC. This early plant breeding was developed primarily to provide a steady supply of food.
18.9K
02:16Light Acquisition
8.5K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
8.5K