CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks
Abstract Background Deep learning techniques have been successfully applied to bioimaging problems; however, these methods are highly data demanding. An approach to deal with the lack of data and avoid overfitting is the application of data augmentation, a technique that generates new training sampl...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2019-06-01
|
Series: | BMC Bioinformatics |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s12859-019-2931-1 |
id |
doaj-6d20978a08e24b708bfacfad33e8202b |
---|---|
record_format |
Article |
spelling |
doaj-6d20978a08e24b708bfacfad33e8202b2020-11-25T02:26:48ZengBMCBMC Bioinformatics1471-21052019-06-0120111410.1186/s12859-019-2931-1CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasksÁngela Casado-García0César Domínguez1Manuel García-Domínguez2Jónathan Heras3Adrián Inés4Eloy Mata5Vico Pascual6Department of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaDepartment of Mathematics and Computer Science, University of La RiojaAbstract Background Deep learning techniques have been successfully applied to bioimaging problems; however, these methods are highly data demanding. An approach to deal with the lack of data and avoid overfitting is the application of data augmentation, a technique that generates new training samples from the original dataset by applying different kinds of transformations. Several tools exist to apply data augmentation in the context of image classification, but it does not exist a similar tool for the problems of localization, detection, semantic segmentation or instance segmentation that works not only with 2 dimensional images but also with multi-dimensional images (such as stacks or videos). Results In this paper, we present a generic strategy that can be applied to automatically augment a dataset of images, or multi-dimensional images, devoted to classification, localization, detection, semantic segmentation or instance segmentation. The augmentation method presented in this paper has been implemented in the open-source package CLoDSA. To prove the benefits of using CLoDSA, we have employed this library to improve the accuracy of models for Malaria parasite classification, stomata detection, and automatic segmentation of neural structures. Conclusions CLoDSA is the first, at least up to the best of our knowledge, image augmentation library for object classification, localization, detection, semantic segmentation, and instance segmentation that works not only with 2 dimensional images but also with multi-dimensional images.http://link.springer.com/article/10.1186/s12859-019-2931-1Data augmentationClassificationDetectionSegmentationMulti-dimensional images |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ángela Casado-García César Domínguez Manuel García-Domínguez Jónathan Heras Adrián Inés Eloy Mata Vico Pascual |
spellingShingle |
Ángela Casado-García César Domínguez Manuel García-Domínguez Jónathan Heras Adrián Inés Eloy Mata Vico Pascual CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks BMC Bioinformatics Data augmentation Classification Detection Segmentation Multi-dimensional images |
author_facet |
Ángela Casado-García César Domínguez Manuel García-Domínguez Jónathan Heras Adrián Inés Eloy Mata Vico Pascual |
author_sort |
Ángela Casado-García |
title |
CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
title_short |
CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
title_full |
CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
title_fullStr |
CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
title_full_unstemmed |
CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
title_sort |
clodsa: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks |
publisher |
BMC |
series |
BMC Bioinformatics |
issn |
1471-2105 |
publishDate |
2019-06-01 |
description |
Abstract Background Deep learning techniques have been successfully applied to bioimaging problems; however, these methods are highly data demanding. An approach to deal with the lack of data and avoid overfitting is the application of data augmentation, a technique that generates new training samples from the original dataset by applying different kinds of transformations. Several tools exist to apply data augmentation in the context of image classification, but it does not exist a similar tool for the problems of localization, detection, semantic segmentation or instance segmentation that works not only with 2 dimensional images but also with multi-dimensional images (such as stacks or videos). Results In this paper, we present a generic strategy that can be applied to automatically augment a dataset of images, or multi-dimensional images, devoted to classification, localization, detection, semantic segmentation or instance segmentation. The augmentation method presented in this paper has been implemented in the open-source package CLoDSA. To prove the benefits of using CLoDSA, we have employed this library to improve the accuracy of models for Malaria parasite classification, stomata detection, and automatic segmentation of neural structures. Conclusions CLoDSA is the first, at least up to the best of our knowledge, image augmentation library for object classification, localization, detection, semantic segmentation, and instance segmentation that works not only with 2 dimensional images but also with multi-dimensional images. |
topic |
Data augmentation Classification Detection Segmentation Multi-dimensional images |
url |
http://link.springer.com/article/10.1186/s12859-019-2931-1 |
work_keys_str_mv |
AT angelacasadogarcia clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT cesardominguez clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT manuelgarciadominguez clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT jonathanheras clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT adrianines clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT eloymata clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks AT vicopascual clodsaatoolforaugmentationinclassificationlocalizationdetectionsemanticsegmentationandinstancesegmentationtasks |
_version_ |
1724845645045432320 |