Ayurvedic Medicinal Raw Materials

The AyuFlora dataset is a high-quality image collection curated to support computer vision and deep learning research in the field of Ayurveda. It is specifically designed to aid in the identification and classification of Ayurvedic raw materials used in medicinal formulations. All images are captured under consistent, controlled lighting conditions to ensure high clarity, contrast, and background uniformity—making the dataset highly suitable for training reliable machine learning models.

Dataset Composition:

The dataset consists of more than 3,000 high-resolution images, divided into 10 distinct classes, each corresponding to a unique Ayurvedic raw material:

Vekhand

Manjistha

Jesthamadh

Gulvel

Ashwagandha

Sagargota

Niranjanphal

Madanphal

Kadukharik

Hirda

Each class contains more than 250 manually curated images—with some classes having 300 or more—ensuring balanced and robust representation across categories for effective model training and evaluation.

Image Characteristics:

Resolution: Typically 224x224 or higher

Lighting Conditions: Consistent and natural/artificial light to enhance visibility

Background: Uniform and noise-free to optimize feature extraction

Format: JPEG/PNG

Applications:

This dataset is ideal for a range of applications, including:

Image classification and recognition of herbal materials

Real-time raw material identification in IoT-enabled systems

Augmented reality support for Ayurvedic practitioners

Transfer learning involving botanical and natural product classification

Additional Files:

A CSV metadata file (ayuflora_dataset.csv) is included with the dataset, containing the following fields for each image: Image_ID, Class_Name, File_Path, and Format.