Biomedical image augmentation using Augmentor

By Marcus D. Bloice, Peter M. Roth, Andreas Holzinger. 2019

225 citations recorded by the source corpus.

MOTIVATION Image augmentation is a frequently used technique in computer vision and has been seeing increased interest since the popularity of deep learning. Its usefulness is becoming more and more recognised due to deep neural networks requiring larger amounts of data to train, and because in certain fields, such as biomedical imaging, large amounts of labelled data are difficult to come by or expensive to produce. In biomedical imaging, features specific to this domain need to be addressed. RESULTS Here we present the Augmentor software package for image augmentation. It provides a stochastic, pipeline-based approach to image augmentation with a number of features that are relevant to biomedical imaging, such as z-stack augmentation and randomised elastic distortions. The software has been designed to be highly extensible, meaning an operation that might be specific to a highly specialised task can easily be added to the library, even at runtime. Although it has been designed as a general software library, it has features that are particularly relevant to biomedical imaging and the techniques required for this domain. AVAILABILITY Augmentor is a Python package made available under the terms of the MIT licence. Source code can be found on GitHub under https://github.com/mdbloice/Augmentor and installation is via the pip package manager*. SUPPLEMENTARY INFORMATION The GitHub repository contains supplementary information, code examples, and Jupyter notebooks. Extensive documentation is hosted on Read the Docs under https://augmentor.readthedocs.io. For continuous integration tests see https://travis-ci.org/mdbloice/Augmentor.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lenses (2020, background): “While earlier computer vision works focused on performing afﬁne transformations to existing samples (O’Gorman & Kasturi, 1995; Bloice et al., 2019), using GANs for synthetic data generation has recently become popular (Teich, 2019; Nisselson, 2018).”
AutoML: A Survey of the State-of-the-Art (2019, background): “Python libraries such as torchvision 5, Augmentor [43], and Albumentations [44] perform these augmentation operations to generate more images.”
Albumentations: fast and flexible image augmentations (2018, methodology): “Over the past few years, multiple image augmentations libraries have been developed, including imgaug [20], torchvision [13], Augmentor [21], CLoDSA [22], SOLT [56], and Automold [57].”
Towards multi-modal causability with Graph Neural Networks enabling information fusion for explainable AI (2021, background): “However, this comes with the downside of decreased generality in the learned representations (since those are now task-specific and thus not reusable across applications), unless informed data augmentation [37] techniques are used.”
Lung and Colon Cancer Histopathological Image Dataset (LC25000) (2019, methodology): “[9] Using Augmentor, we expanded our dataset to 25,000 images by the following augmentations: left and right rotations (up to 25 degrees, 1.0 probability) and by horizontal and vertical flips (0.5 probability).”
TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning (2020, unknown): “e is loaded from disk during training. Multiple libraries with support for computer vision data augmentation have appeared in the last years, such as Albumentations [Buslaev et al., 2020], Augmentor [Bloice et al., 2019], Kornia [Riba et al., 2019], or imgaug [Jung et al., 2020]. PyTorch also includ”
Democratising deep learning for microscopy with ZeroCostDL4Mic (2021, methodology): “Important packages used in ZeroCostDL4Mic include Numpy 76 , Keras (https://github.com/keras-team/keras), TensorFlow 77 , Augmentor 32 , tiff ﬁ le (https://github.com/cgohlke/tiff ﬁ le), elasticdeform 28 and imgaug.”
TrackMate 7: integrating state-of-the-art segmentation algorithms into tracking pipelines (2022, methodology): “The generated training source and target were randomly cropped into size 1024 × 1024, rotated, flipped, and multiplied by 5 using the Augmentor ZeroCostDL4Mic notebook 15,29 to generate a dataset of 120 paired images.”