This article presents a curated medicinal plant leaf image dataset designed to support research in plant health monitoring and computer vision applications. The dataset contains images from five medicinal plant species: Aloe Vera, Azadirachta Indica (Neem), Hibiscus Rosa Sinensis, Kalanchoe Pinnata, and Piper Betle. Leaf images were collected between January 7, 2026 and February 27, 2026 from Rajbari, Ashulia, and Mirpur in Dhaka, Bangladesh. Images were captured individually using smartphone cameras (OnePlus Nord CE 4 Lite and iPhone 16 Pro Max) under controlled acquisition conditions where detached leaves were placed on a uniform background to improve visibility of leaf morphology and surface symptoms. The dataset includes 1,323 original images representing 16 leaf condition classes, including healthy, chlorotic, diseased, and dried. During preprocessing, images were standardized by resizing them to 512 × 512 pixels, removing background regions, and normalizing pixel values to ensure consistent input for machine learning pipelines. Data augmentation techniques such as rotation, horizontal and vertical flipping, brightness and contrast adjustment, Gaussian noise addition, and image sharpening were applied to generate additional image samples and increase dataset variability, producing a processed dataset of 14,677 images. The dataset is organized into original and processed image directories and is accompanied by a CSV metadata file describing plant species, leaf condition labels, image counts, and collection locations. These data can be reused for image classification, plant disease detection, agricultural computer vision, and machine learning model development. Keywords: Medicinal plant dataset, Plant disease detection, Computer vision, Machine learning, Leaf image dataset
Hossan et al. (Sat,) studied this question.