SAR-MLD1-2025-MangoLeaf: A comprehensive high-quality mango leaf dataset for disease classification

Area of Science:

• Agricultural Science
• Computer Science
• Botany

Background:

• Mango production in Bangladesh is significantly impacted by leaf diseases, affecting yield and quality.
• Early detection of mango leaf diseases is critical for effective management and preventing widespread crop damage.
• Automated disease detection systems can enhance agricultural practices and improve disease management strategies.

Purpose of the Study:

• To present a comprehensive dataset of mango leaf images for research purposes.
• To facilitate the development of machine learning models for automated mango leaf disease classification.
• To provide a valuable resource for botanical and agricultural research in Bangladesh.

Main Methods:

• Collected 4921 raw image samples of mango leaves from Naogaon District, Bangladesh.
• Categorized images into five classes: Healthy, Anthracnose, Powdery Mildew, Turning Brown, and Gall Midge.
• Captured images under natural lighting conditions to ensure authenticity and variability.

Main Results:

• A curated dataset of 4921 labeled mango leaf images is now available.
• The dataset includes diverse visual representations of healthy and diseased mango leaves.
• The images captured under natural light preserve intrinsic visual features for robust analysis.

Conclusions:

• The presented dataset is a significant resource for advancing research in automated mango leaf disease detection.
• This dataset will aid researchers in developing and validating machine learning models for improved disease diagnosis.
• Utilizing this dataset can lead to more effective disease management strategies, benefiting mango cultivation in Bangladesh and beyond.