Related Experiment Video
Updated: Jan 14, 2026
06:41
Imaging and Analysis for Quantifying Maize (Zea mays) Abiotic Stress Phenotypes
Published on: March 28, 2025
1.5K
A comprehensive annotated image dataset for deep learning analysis of eggplant leaf diseases
Md Asraful Sharker Nirob1, Prayma Bishshash1, Mariyam Bin Ayan1
1Health Informatics Research Lab, Department of CSE, Daffodil International University, Daffodil Smart City, Birulia, Savar, Dhaka 1216, Bangladesh.
Data in Brief
|October 27, 2025
Summary
A new eggplant leaf disease dataset and a CBAM-EfficientNetB0 model achieve 98.70% accuracy for early disease detection. This AI-powered approach aids precision agriculture, improving crop yields and food security.
Area of Science:
- Agricultural Science
- Computer Science
- Plant Pathology
Background:
- Eggplant diseases pose a significant threat to global food security.
- Accurate and early disease identification is crucial for effective crop management.
Purpose of the Study:
- To develop a comprehensive dataset for eggplant leaf disease identification.
- To propose and evaluate a novel deep learning model for accurate disease classification.
Main Methods:
- Collected 3116 high-resolution images of eggplant leaves from Bangladesh.
- Expanded the dataset to 10,000 images using data augmentation techniques.
- Developed and trained a Convolutional Block Attention Module-EfficientNetB0 (CBAM-EfficientNetB0) model.
Main Results:
- The proposed CBAM-EfficientNetB0 model achieved a classification accuracy of 98.70%.
- Outperformed baseline models like ResNet50 (32.60%), VGG16 (73.00%), and VGG19 (68.00%).
- The model effectively extracts distinguishing features for precise disease identification.
Conclusions:
- The developed dataset and CBAM-EfficientNetB0 model enable AI-powered early disease detection in eggplants.
- This technology supports automated monitoring and decision-making in precision agriculture.
- Facilitates sustainable farming practices, increased crop productivity, and enhanced food security.
