BanglaLeaf: Leaf Image Dataset of Locally Grown Gourd Varieties in Bangladesh for Leaf-Based Species Classification
Description
This dataset contains leaf images of six gourd varieties (family Cucurbitaceae) commonly grown in Bangladesh: Sponge Gourd (ধুন্দল, Luffa cylindrica), Pumpkin (মিষ্টিকুমড়া, Cucurbita moschata), Ash Gourd (চালকুমড়া, Benincasa hispida), Bottle Gourd (লাউ, Lagenaria siceraria), Bitter Gourd (করলা, Momordica charantia), and Little Gourd (তেলাকুচা, Coccinia grandis). These species are widely cultivated across Bangladeshi home gardens, small farms, and local markets and are staple vegetables in regional cuisine. A total of 1,308 original images were collected: 240 Sponge Gourd, 218 Pumpkin, 200 Ash Gourd, 217 Bottle Gourd, 204 Bitter Gourd, and 229 Little Gourd. Photos were taken using an iPhone 13 and an iPhone 6s Plus, captured in HEIC format and converted to JPG using DocuFreezer. Meanwhile, the dataset is expanded to 6000 images through data augmentation, including rotation, flipping, brightness, contrast, color variation, and slight zoom, applied to increase sample diversity for model training. The images are provided in JPG format at 3024 x 4032 pixels and 96 dpi resolution. Original and augmented images are organized separately into class-wise folders, allowing researchers to work with the raw images alone or use the expanded set directly for training purposes. This dataset is useful for academics and researchers working in agriculture, plant classification, computer vision, and image processing. It can be used to create and evaluate computer vision models for image classification, detection, and other related tasks. It may also serve as a regional resource for studies on gourd leaf morphology and species identification in Bangladesh. Dataset Overview: - 1,308 original leaf images with 6 different varieties - 6000 augmented leaf images - JPG image format - High-quality RGB photos, 3024 x 4032 pixels, 96 dpi
Files
Institutions
- Daffodil International UniversityDhaka Division, Dhaka