Related Experiment Video
Updated: May 20, 2025
05:56
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
Published on: April 14, 2023
2.3K
A data-driven approach to turmeric disease detection: Dataset for plant condition classification
A K M Fazlul Kobir Siam1, Md Asraful Sharker Nirob1, Prayma Bishshash1
1Department of CSE, Daffodil International University, Bangladesh.
Data in Brief
|March 27, 2025
Summary
This study introduces an AI model for early turmeric plant disease diagnosis using image classification. The model achieved 97.36% accuracy, aiding precision farming and sustainable agriculture.
Area of Science:
- Agricultural Science
- Computer Science
- Plant Pathology
Background:
- Turmeric (Curcuma longa) is vital economically and medicinally but susceptible to diseases like leaf blotch and rhizome rot.
- Current disease diagnosis methods are manual, time-consuming, subjective, and impractical for large-scale farming.
- Accurate and early disease detection is crucial for minimizing crop losses and promoting sustainable agriculture.
Purpose of the Study:
- To develop and evaluate an AI-based system for accurate and efficient turmeric plant disease classification.
- To create and share a comprehensive dataset of turmeric plant images for deep learning applications.
- To demonstrate the potential of artificial intelligence in enhancing precision farming and sustainable crop management.
Main Methods:
- A dataset of 1037 original and 4628 augmented images of turmeric plants (healthy, dry leaf, leaf blotch, rhizome disease roots, healthy roots) was curated.
- Image data underwent pre-processing including resizing, cleaning, and augmentation (flipping, rotation, brightness adjustment).
- The Inception-v3 deep learning model was employed for plant disease classification.
Main Results:
- The Inception-v3 model achieved 97.36% accuracy with data augmentation, outperforming the 95.71% accuracy without augmentation.
- Key performance metrics such as precision, recall, and F1-score confirmed the model's robustness and efficacy.
- The study highlights the effectiveness of AI in differentiating between healthy and diseased turmeric plant samples.
Conclusions:
- AI-driven image classification offers a promising solution for early and accurate diagnosis of turmeric plant diseases.
- The developed dataset and model contribute to advancing AI applications in precision agriculture and sustainable crop production.
- This research encourages further investigation into AI-powered tools for agricultural disease management, particularly in developing regions.
