{"full_name":"aleju/imgaug","owner":"aleju","name":"imgaug","shard":"1","url":"https://github.com/aleju/imgaug","description":"Image augmentation for machine learning experiments.","homepage":"http://imgaug.readthedocs.io","topics":["image-augmentation","machine-learning","augmentation","deep-learning","images","affine-transformation","polygon","heatmap","segmentation-maps","keypoints","bounding-boxes","augment-images","crop","contrast"],"tag_pages":["augmentation","contrast","crop","deep-learning","image-augmentation","images","machine-learning"],"language":"Python","license":"MIT","avatar":"https://avatars.githubusercontent.com/u/11698516?v=4","created_at":"2015-07-10T20:31:33Z","pushed_at":"2024-07-30T01:38:33Z","metrics":{"stars":14739,"forks":2445,"watchers":221,"issues":519,"prs":352,"releases":3,"issues_open":290,"issues_closed":229,"prs_open":21,"prs_closed":18,"prs_merged":313,"discussions":0},"growth":{"day":{"stars":0,"forks":0,"base_day":"2026-11-15"},"week":{"stars":5,"forks":2,"base_day":"2026-11-04"},"month":{"stars":-5,"forks":2,"base_day":"2026-10-17"}},"history":{"dates":["2026-09-06","2026-09-07","2026-09-08","2026-09-09","2026-09-10","2026-09-11","2026-09-12","2026-09-13","2026-09-14","2026-09-15","2026-09-16","2026-09-17","2026-09-18","2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-10","2026-11-11","2026-11-12","2026-11-13","2026-11-14","2026-11-15","2026-11-16"],"stars":[14739,14740,14740,14736,14734,14735,14734,14735,14735,14734,14733,14736,14737,14737,14736,14736,14738,14740,14741,14741,14741,14741,14740,14740,14741,14741,14742,14744,14744,14744,14744,14744,14745,14745,14744,14744,14744,14744,14745,14745,14745,14744,14741,14740,14740,14738,14738,14738,14738,14735,14736,14735,14735,14735,14735,14735,14735,14735,14734,14734,14738,14741,14739,14739,14739,14739,14739],"forks":[2450,2451,2451,2451,2451,2451,2450,2450,2450,2450,2450,2450,2450,2450,2450,2450,2450,2449,2449,2449,2449,2448,2448,2448,2446,2447,2447,2447,2447,2447,2445,2445,2445,2446,2446,2446,2446,2446,2446,2445,2443,2443,2443,2443,2443,2443,2443,2443,2443,2443,2443,2443,2443,2444,2444,2444,2443,2443,2443,2443,2444,2444,2445,2445,2445,2445,2445],"watchers":[222,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,223,221,221,221,221,221,221,221,221,221,221,221,null,221,221,221,221,221,221],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,290,290,290,290,290,290],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,21,21,21,21,21,21]},"boards":{},"same_owner":[],"issues":[{"number":473,"title":"Explicit OpenCV flavor requirement conflicts with existing installs of OpenCV","comments":20,"created_at":"2019-11-01T15:12:02Z"},{"number":66,"title":"Using this library with Keras","comments":20,"created_at":"2017-09-29T09:42:40Z"},{"number":79,"title":"Using OpenCV's warpAffine","comments":19,"created_at":"2017-11-19T18:24:04Z"},{"number":717,"title":"Segmentation Aug with Labelme Json files?","comments":17,"created_at":"2020-09-24T20:27:35Z"},{"number":310,"title":"Bug in clip_out_of_image","comments":13,"created_at":"2019-05-09T16:35:43Z"},{"number":35,"title":"Rotation, which changes shape","comments":13,"created_at":"2017-05-25T13:51:40Z"},{"number":56,"title":"Augmentation by Perspective Transform","comments":12,"created_at":"2017-08-15T17:39:24Z"},{"number":406,"title":"how to use imgaug with pytorch","comments":11,"created_at":"2019-08-30T07:17:43Z"}],"discussions":[],"latest_release":{"name":"0.4.0","tag":"0.4.0","published_at":"2020-02-06T06:18:40Z","body":"# Table of Contents\r\n\r\n1. [Overview](#overview)\r\n2. [Example Images](#example_images)\r\n3. [Mixed-Category Patches](#mixed_category_patches)\r\n4. [Added](#added)\r\n5. [Changed](#changed)\r\n6. [Refactored](#refactored)\r\n7. [Fixed](#fixed)\r\n\r\n\r\n<a name=\"overview\"/>\r\n\r\n# Overview\r\n\r\nRelease `0.4.0` focused mainly on adding new augmenters and improving\r\nthe internal augmentation \"backend\".\r\n\r\nThe following augmenters were added (see the [overview](https://imgaug.readthedocs.io/en/latest/source/overview_of_augmenters.html) docs for more details):\r\n* `ChangeColorTemperature`: Gives images a red, orange or blue touch.\r\n* New Brightness augmenters: `WithBrightnessChannels`,\r\n  `MultiplyAndAddToBrightness`, `MultiplyBrightness`, `AddToBrightness`.\r\n* New Dropout augmenters: `Dropout2d`, `TotalDropout`.\r\n* `RemoveSaturation`: Decreases the saturation of colors. Effects are similar\r\n  to `Grayscale`.\r\n* `Cartoon`: Applies a cartoon-style to images (classical / not-learned).\r\n* `MeanShiftBlur`: Blurs images using a mean-shift clustering method.\r\n  (Note: Very slow.)\r\n* `Jigsaw`: Splits the image into rectangular cells and randomly switches\r\n  some pairs of neighbouring cells. (Note: Does not support bounding boxes,\r\n  polygons and line strings.)\r\n* `WithPolarWarping`: Transforms images to polar coordinate space and applies\r\n  child augmenters there.\r\n* `SaveDebugImageEveryNBatches`: Generates and saves at every `N`-th batch a\r\n  debug image visualizing all inputs within the batch. Useful to gauge\r\n  strength and effects of augmentations and quickly spot errors in ground truth\r\n  data (e.g. misaligned bounding boxes).\r\n* `Cutout`: Removes rectangular subregions of images. Has some similarity with\r\n  `CoarseDropout`.\r\n* `Rain` and `RainLayer`: Adds rain-like effects to images.\r\n* `RandAugment`: Combination of multiple augmenters. Similar to the paper\r\n  description. (Note: Can currently only augment images.)\r\n* `Identity`: Same as `Noop`. Does nothing.\r\n* `UniformColorQuantizationToNBits`: Quantizes each image array component down\r\n  to `N` bits. Similar to `UniformColorQuantization`. Has the alias\r\n  `Posterize`.\r\n* `Solarize`: Invert with threshold.\r\n* `RemoveCBAsByOutOfImageFraction`, `ClipCBAsToImagePlanes`: Augmenters to\r\n  remove or clip coordinate-based augmentables, e.g. bounding boxes\r\n* More blend augmenters:\r\n  * `BlendAlphaMask`: Uses batch-wise generated masks for alpha-blending.\r\n  * `BlendAlphaSomeColors`: Alpha-blends only within image regions having\r\n    specific randomly chosen colors.\r\n  * `BlendAlphaSegMapClassIds`: Alpha-blends only within image regions having\r\n    specific class ids in segmentation maps.\r\n  * `BlendAlphaBoundingBoxes`: Alpha-blends only within image regions covered\r\n    by bounding boxes having specific labels.\r\n  * `BlendAlphaHorizontalLinearGradient`: Alpha-blends using horizontal\r\n    linear gradients.\r\n  * `BlendAlphaVerticalLinearGradient`: Analogous.\r\n  * `BlendAlphaRegularGrid`: Places a regular grid on each image and\r\n    samples one alpha value per grid cell. Can be used e.g. to achieve\r\n    coarse dropout.\r\n  * `BlendAlphaCheckerboard`: Places also a regular grid on each image,\r\n    but neighbouring cells use alpha values that are inverse to each other.\r\n* Shortcuts for `Affine`: `ScaleX`, `ScaleY`, `TranslateX`, `TranslateY`,\r\n  `Rotate`, `ShearX`, `ShearY`.\r\n* Several new crop and pad augmenters: `CenterCropToFixedSize`,\r\n  `CenterPadToFixedSize`, `CropToMultiplesOf`, `CenterCropToMultiplesOf`,\r\n  `PadToMultiplesOf`, `CenterPadToMultiplesOf`, `CropToPowersOf`,\r\n  `CenterCropToPowersOf`, `PadToPowersOf`, `CenterPadToPowersOf`,\r\n  `CropToAspectRatio`, `CenterCropToAspectRatio`, `PadToAspectRatio`,\r\n  `CenterPadToAspectRatio`, `PadToSquare`, `CenterPadToSquare`,\r\n  `CropToSquare`, `CenterCropToSquare`.\r\n* Wrappers around `imagecorruptions` package (verified to have identical\r\n  outputs):\r\n  `GaussianNoise`, `ShotNoise`, `ImpulseNoise`, `SpeckleNoise`,\r\n  `GaussianBlur`, `GlassBlur`, `DefocusBlur`,","is_prerelease":false},"readme":true,"readme_zh":false,"news":[],"stale":false,"snapshot_at":"2026-11-16T09:23:14Z","history_start":"2026-09-06"}