MedicinalLeafBD: A complete image dataset for classification of medicinal plant leaves found in Bangladesh

The MedicinalLeafBD dataset is a comprehensive collection of 8,500 high-quality images of medicinal plant leaves, systematically gathered from the National Botanical Garden of Bangladesh, Mirpur, Dhaka 1216; Noakhali Nursery, Uttara, Dhaka; Moulvibazar, Sylhet; and rural regions of Bhola Sadar, Bangladesh, between March 5, 2025 to June 15, 2025. It consists of 20 distinct classes commonly used in traditional and herbal medicine practices across Bangladesh, with the following distribution: Ashok (437), Akondo (426), Basil (461), Cannonball Tree (424), Cordia (426), Debdaru (416), Ginger (408), HoludPata (406), Jarul (414), Mastwood (419), Minjiri (443), Nageshor (419), Neem (446), Pathorkuchi (434), Punnag (427), Royna (411), Tejpata (405), Telakucha (429), Thankuni (433), and Vasaka (416). All images were captured using an iPhone 15 under natural lighting conditions and standardized to 640 × 480 pixels at 96 dpi. This dataset offers a valuable resource for botanical studies, pharmacological research, biodiversity documentation, and AI-driven plant recognition systems, contributing to healthcare innovation, sustainable agriculture, and environmental conservation.