140 Most Popular Crops Image Dataset
Complete image dataset of 139 agricultural crops, with approximately 250 images per class, used for recognition and classification in agriculture.
~104,000 RGB, BGR and greyscale images, 224x224 pixel resolution
CC0: Public Domain
Description
This dataset includes approximately 104,000 images divided into 139 classes representing popular agricultural crops. Each class contains approximately 250 images, available in RGB, BGR (OpenCV compatible) and gray scale formats, with a standard resolution of 224x224 pixels.
What is this dataset for?
- Train models for the recognition and classification of agricultural plants and crops
- Develop precision agriculture solutions and AI crop monitoring
- Testing multi-format and multi-channel adaptation for computer vision
Can it be enriched or improved?
Yes, it is possible to add additional annotations, such as the detection of diseases or growth stages, or to integrate contextual metadata (location, season).
🔎 In summary
🧠 Recommended for
- AI agricultural researchers
- Computer vision students
- Classification model developers
🔧 Compatible tools
- TensorFlow
- PyTorch
- OpenCV
- FastAI
💡 Tip
Take advantage of the three image formats to experiment with robust multi-channel models.
Frequently Asked Questions
Is this dataset suitable for the detection of diseases on crops?
No, it is primarily designed for the classification of crop types, without disease annotation.
Can this dataset be used to train a multi-channel model?
Yes, the images are provided in RGB, BGR and gray levels, allowing multi-format approaches to be explored.
What is the standard image resolution?
All images are standardized at 224x224 pixels, adapted to most standard CNN architectures.
