Multi-Source Agricultural Image Dataset for Robust Classification and Semantic Segmentation

This dataset includes images of various agricultural items, grouped into four main categories: crops and grains, flowers, fruits, and vegetables. Each folder contains subcategories named using standard, widely understood English terms to support easy use in machine learning tasks.

In total, there are over 4,900 .jpg images, each resized to 512×512 pixels. Every subcategory includes 70 images, making the dataset consistent and balanced.

Highlights:

--4 primary categories

--70+ subcategories

--All images in .jpg, sized to 512×512

--Labels use consistent English naming

Use Cases:

--Training image classifiers for fruits, vegetables, and flowers

--Supporting agricultural and botanical research

--Educational projects or resource-efficient AI systems

This dataset is suitable for both research and practical applications in agriculture and Arttificial Intelligence.