Meta_Album_PLK_Mini
OpenML44282MaRDI QIDQ6037277FDOQ6037277RO-CrateQ6037277
OpenML dataset with id 44282
Ihsan Ullah
Full work available at URL: https://api.openml.org/data/v1/download/22110982/Meta_Album_PLK_Mini.arff
Upload date: 28 October 2022
Copyright license: Creative Commons Attribution-NonCommercial 4.0 International
Dataset Characteristics
Number of classes: 86
Number of features: 3 (numeric: 1, symbolic: 0 and in total binary: 0 )
Number of instances: 3,440
Number of instances with missing values: 3,440
Number of missing values: 3,440
Meta-Album Plankton Dataset (Mini)
* The Plankton dataset is created by researchers at the Woods Hole Oceanographic Institution (https://www.whoi.edu/). Imaging FlowCytobot (IFCB) was used for the data collection. The Complete process and mechanism are described in the paper 31. Each image in the dataset contains one or multiple planktons. The images are captured in a controlled environment and have different orientations based on the flow of the fluid in which the images are captured and the size and shape of the planktons. The preprocessed plankton dataset is prepared from the original WHOI Plankton dataset. The preprocessing of the images is done by creating a background squared image by either duplicating the top and bottom-most 3 rows or the left and right most 3 columns based on the orientation of the original image to match the width or height of the image respectively. A Gaussian kernal of size 29x29 is applied to the background image to blur the image. Finally, the original plankton image is pasted on the background image at the center of the image. The squared background image with the original plankton image on top of it as one image is then resized into 128x128 with anti-aliasing.
Dataset Details
![]
Meta Album ID: SM_AM.PLK Meta Album URL: https://meta-album.github.io/datasets/PLK.html Domain ID: SM_AM Domain Name: Small Animals Dataset ID: PLK Dataset Name: Plankton Short Description: Plankton dataset from WHOI \# Classes: 86 \# Images: 3440 Keywords: plankton, ecology Data Format: images Image size: 128x128
License (original data release): MIT License License URL(original data release): https://github.com/hsosik/WHOI-Plankton/blob/master/LICENSE
License (Meta-Album data release): MIT License License URL (Meta-Album data release): https://github.com/hsosik/WHOI-Plankton/blob/master/LICENSE
Source: Woods Hole Oceanographic Institution Source URL: https://github.com/hsosik/WHOI-Plankton
Original Author: Heidi M. Sosik, Emily E. Peacock, Emily F. Brownlee, Eric Orenstein Original contact: hsosik@whoi.edu
Meta Album author: Ihsan Ullah Created Date: 01 March 2022 Contact Name: Ihsan Ullah Contact Email: meta-album@chalearn.org Contact URL: https://meta-album.github.io/
Cite this dataset
``` @misc{whoiplankton,
title={Annotated Plankton Images - Data Set for Developing and Evaluating Classification Methods.},
author={Heidi M. Sosik, Emily E. Peacock, Emily F. Brownlee},
year={2015},
DOI = {10.1575/1912/7341},
url={https://hdl.handle.net/10.1575/1912/7341}
} ```
Cite Meta-Album
``` @inproceedings{meta-album-2022,
title={Meta-Album: Multi-domain Meta-Dataset for Few-Shot Image Classification},
author={Ullah, Ihsan and Carrion, Dustin and Escalera, Sergio and Guyon, Isabelle M and Huisman, Mike and Mohr, Felix and van Rijn, Jan N and Sun, Haozhe and Vanschoren, Joaquin and Vu, Phan Anh},
booktitle={Thirty-sixth Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
url = {https://meta-album.github.io/},
year = {2022}
}
```
More
For more information on the Meta-Album dataset, please see the [NeurIPS 2022 paper] For details on the dataset preprocessing, please see the [supplementary materials] Supporting code can be found on our [GitHub repo] Meta-Album on Papers with Code [Meta-Album]
Other versions of this dataset**
ROCrate
What is a RO-Crate?
A RO-Crate is a standardized research object package used to bundle data together with rich machine-readable metadata. Each RO-Crate contains:
- the files belonging to the dataset (e.g. CSVs, images, code, documentation)
- a ro-crate-metadata.json file describing the content, provenance, and context
- persistent identifiers and references to related research objects (e.g. software, publications)
This ensures that the dataset can be easily reused, cited, validated, and interpreted in a reproducible manner. More information can be found here.
Download
You can download a RO-Crate for this dataset here: Download RO-Crate
HINT: The RO-Crate is created dynamically, so it could take up to 30 seconds until the downloads starts.
This page was built for dataset: Meta_Album_PLK_Mini
