Skip to content

Latest commit

 

History

History
44 lines (37 loc) · 1.64 KB

README.md

File metadata and controls

44 lines (37 loc) · 1.64 KB

INHS Fish Outlining

Example fish outline

Setup

1.

Ensure you have the latest version of fish.db. I've reached my free Git LFS quota, so the copy included here may be old. The latest version has this SHA1 hash:

$ sha1sum fish.db
7a047498773f207c17fbecf0d63fbd9365f973b5  fish.db

If your version differs, please download the latest:

wget http://andrewsenin.com/fish.db

2.

Create a Python 3 virtual environment in the top level of the repository:

$ python -m venv venv/
$ source venv/bin/activate
(venv) $ pip install -r requirements.txt

Reproducing presentation results

1.

Generate the dataset. This shouldn't take more than one half hour on an average processor:

python datagen.py

This produces the following dataset files in the current directory:

  • 1mm_fifteen_species.csv
  • 1mm_seven_genera.csv
  • 1mm_aug_seven_genera.csv

2.

Run the classification experiment by loading your desired dataset in classification.ipynb and running all the cells.

Querying and viewing fish

See outlining_demo.ipynb for usage examples.

Associated publications

J. Pepper, J. Greenberg, Y. Bakiş, X. Wang, H. Bart and D. Breen, "Automatic Metadata Generation for Fish Specimen Image Collections," 2021 ACM/IEEE Joint Conference on Digital Libraries (JCDL), 2021, pp. 31-40, doi: 10.1109/JCDL52503.2021.00015.

Kevin Karnani, Joel Pepper, Yasin Bakis et al. Computational Metadata Generation Methods for Biological Specimen Image Collections, 27 April 2022, PREPRINT (Version 1) available at Research Square https://doi.org/10.21203/rs.3.rs-1506561/v1