Correction to: Scientific Data https://doi.org/10.1038/sdata.2018.172, published online 28 August 2018
Following further analysis of the Majority Dataset (Data Citation 3, originally https://doi.org/10.23728/b2share.e344a8afef08463a855ada08aadbf352) and 100% Dataset (Data Citation 4, originally https://doi.org/10.23728/b2share.f1aa0f5ad38c456eaf7b04d47a65af53) presented in the original version of this Data Descriptor it was revealed that a large number of duplicate images were included in both datasets. Both datasets have been corrected in updated versions, removing all replicates. The new version of the Majority Dataset (Data Citation 3) can be accessed via https://doi.org/10.23728/b2share.72758204db9044ab8b3e6b6c4d2eb576 and the 100% Dataset (Data Citation 4) via https://doi.org/10.23728/b2share.80df8606fcdb4b2bae1656f0dc6db8ba. The HTML and PDF versions of the Data Descriptor have been corrected accordingly.
In addition, the authors found that the category ‘3d array’ was erroneously added to the Hierarchical Dataset (Data Citation 2). This has now been removed.
In order to account for the modifications to the datasets described above, a number of textual modifications to the HTML and PDF versions of Data Descriptor have been made. These are listed by their location below:
Abstract
The number of SEM images classified was previously stated as ~26,000. This has been correct to ~22,000.
Background & Summary
Where it was previously stated that the Majority Dataset and 100% Dataset contain ~7,000 newly classified images, the text has been corrected to state ~3,000 newly classified images.
Data Records
The two increased datasets were stated as being obtained by classifying and validating ~7,000 images. This has been corrected to ~3,000 images.
Hierarchical dataset
The number of subcategories into which images were classified has been corrected from 27 to 26.
Majority dataset
The number of SEM images in the Majority Dataset has been corrected from 25,537 to 21,272.
100% dataset
The number of SEM images in the 100% Dataset has been corrected from 25,430 to 21,169.
Technical Validation
Previously it was stated that ~7,000 images were labelled since October 2017, and the size of the dataset correspondingly increased by ~37%. This has been corrected to ~3,000 images and ~17%.
Table 2
Previously Table 2 included a row concerning the ‘3d array’ subcategory that was added erroneously. This has now been removed.
Table 3
The number of categories in the Hierarchical Dataset was corrected from 37 to 36. The number of images in the Majority and 100% datasets was corrected from 25,537 to 21,272 and from 25,430 to 21,169, respectively.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Aversa, R., Modarres, M.H., Cozzini, S. et al. Author Correction: The first annotated set of scanning electron microscopy images for nanoscience. Sci Data 6, 3 (2019). https://doi.org/10.1038/s41597-019-0007-8
Published:
DOI: https://doi.org/10.1038/s41597-019-0007-8
