A comprehensive dataset of garlic leaf images for the identification of Tipburn disease in Bangladesh
Description
The Garlic Leaf Dataset consists of images of Tipburn-affected leaves and healthy leaves. The dataset is organized into two main subfolders: Healthy Leaf and Affected Leaf.Each of these folders is further divided into two subfolders: Original Images and Augmented Images.The Healthy Leaf folder contains 3,454 images and the Affected Leaf folder contains 3,990 images. In total, the garlic leaf dataset contains 7,444 high-quality images. The dataset was collected from different locations across Bangladesh at various times of the day and was verified by an agriculture specialist. It is suitable for applying deep learning and machine learning methods.
Files
Steps to reproduce
To reproduce the dataset, one should first consult an agriculture specialist to gather information about the disease and its characteristics. Afterward, field visits should be conducted to talk with farmers and identify affected areas. Data should be collected based on these observations, and any noise or blurred entries must be removed to maintain quality. To increase dataset size and diversity, appropriate data augmentation methods should be applied. Finally, all collected and augmented data should be verified by the agriculture specialist to ensure accuracy. This process ensures that the dataset is reliable and suitable for deep learning and machine learning applications.
Institutions
- Khwaja Yunus Ali University