aleju/imgaug

Project overview

By covering many augmentation techniques in a single Python library, imgaug lowers the effort required to expand training datasets for image-based machine learning, which is a recurring need for developers and data teams.

Project type

Model Development · Data Processing · Image & Vision

Deployment

Refer to project documentation

License

MIT

Best for

• Developers, data teams, and AI engineers who need a Python library to easily apply many augmentation techniques to image datasets for machine learning projects.

Key capabilities

• Applies many augmentation techniques to input images to convert them into a new, much larger set of slightly altered images.

Limitations and risks

• The library supports Python 2.7 and 3.4+.

Getting started

• Install the library using pip (pip install imgaug) or conda, then import it in Python code using import imgaug.augmenters as iaa to begin applying augmentations.

Evidence and sources

• README: This python library helps you with augmenting images for your machine learning projects.
• GitHub project description: Image augmentation for machine learning experiments.
• README: * Optimized for high performance
• Release: 0.4.0
• README: The library supports python 2.7 and 3.4+.

Reviewed at: 2026-07-17T14:49:30Z