# aleju/imgaug

Image augmentation for machine learning experiments.

Repository: https://github.com/aleju/imgaug
Canonical: https://ross.abutalabs.com/products/imgaug
Homepage: http://imgaug.readthedocs.io
Language: Python
License: MIT
License Family: permissive
Topics: image-augmentation, machine-learning, augmentation, deep-learning, images, affine-transformation, polygon, heatmap, segmentation-maps, keypoints, bounding-boxes, augment-images, crop, contrast
Last push: 2024-07-30T01:38:33+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 4072, "days_push": 765, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 14741, forks 2451 (observed 2026-08-28T04:11:07.588683+00:00)

## What it is
imgaug is a Python library for augmenting images in machine learning experiments, converting a small set of input images into a much larger set of slightly altered versions. It supports a wide range of augmentation techniques and can consistently augment not only images but also keypoints, bounding boxes, heatmaps, and segmentation maps.

## Use cases
- augment images for training deep learning models
- generate more training data from a small image dataset
- augment bounding boxes and keypoints alongside images
- augment segmentation maps and heatmaps consistently with images
- reduce overfitting in CNN training with varied images
- apply random affine transformations, crops, and contrast changes to images

## When to choose
- you need image augmentation that keeps labels like bounding boxes, keypoints, and segmentation maps in sync
- you want a large catalog of ready-made augmenters (weather, blur, contrast, geometric) combinable in pipelines
- you need stochastic, reproducible augmentation with multi-core execution

## When to avoid
- you need GPU-accelerated augmentation tightly integrated with a modern framework like PyTorch or TensorFlow (consider albumentations or built-in pipeline transforms)
- you need active development and new features, as the project is largely in maintenance
- you augment non-image modalities such as text or audio

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, data-generation
- domain: machine-learning, computer-vision, deep-learning, image-processing
- platform: python, cross-platform
- tags: image-augmentation, data-augmentation, bounding-boxes, keypoints, segmentation-maps, heatmaps, computer-vision, deep-learning

## Member repositories
- aleju/imgaug (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:07.588683+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:07:12.946224+00:00, confidence not recorded.
  - readme: https://github.com/aleju/imgaug (fetched 2026-08-28T04:11:07.588683+00:00, sha 196428e1281e)
  - homepage: http://imgaug.readthedocs.io (fetched 2026-08-29T08:05:16.776306+00:00, sha 65b6809a76ee)
  - registry_pypi: https://pypi.org/pypi/imgaug/json (fetched 2026-08-29T08:05:16.785526+00:00, sha e0891ee05569)
- Data as of 2026-08-30T08:39:29.467469+00:00.