Fragrant Flowers Western Region of India

This dataset contains a curated collection of flower images representing multiple botanical species, intended for use in computer vision and machine learning applications such as image classification and plant identification.
The dataset includes commonly found flowering plants such as Jasminum sambac (Mogra), Jasminum auriculatum (Jai/Jui), Jasminum multiflorum (Kunda), Nyctanthes arbor-tristis (Parijat), Tabernaemontana divaricata (White Anant), Plumeria alba (White Chafa), Magnolia champaca (Son Chafa), Cestrum nocturnum (Raatrani), Passiflora caerulea (Krishna Kamal), and Rosa species (Red and Pink Rose).
All images were captured under real-world conditions, incorporating variations in lighting, viewing angles, background complexity, and natural environmental settings to enhance dataset diversity and model generalization.
The dataset is organized into class-wise directories using scientific names to ensure taxonomic consistency. A corresponding metadata file (metadata.csv) is provided, containing filename, class labels (scientific and common names), and a source indicator distinguishing original and augmented samples. Approximately 30% of the data is labeled as augmented, maintaining class-wise balance to support robust model training.
This dataset is suitable for multi-class classification, transfer learning, fine-grained visual recognition, and AI-based plant identification systems, and is designed to be directly usable in standard machine learning workflows.