JAMSTEC E-library
of Deep-sea Images (J-EDI)

  • menu

An Image AI Dataset for Deep-sea Organism Detection

Abstract

This is an AI training dataset for detecting organisms from images taken during deep-sea survey. The detection targets are labeled into 19 rough categories such as "shrimp" and "fish". The dataset consists of over 8,000 images sliced from JAMSTEC's deep-sea survey videos on "JAMSTEC E-library of Deep-sea Images (J-EDI)" and COCO (Common Objects in Context) format JSON including bounding boxes and labels. The dataset can be downloaded as a zip file.

Dataset examples

The dataset was created by using following software:

Abhishek Dutta and Andrew Zisserman. 2019. The VIA Annotation Software for Images, Audio and Video. In Proceedings of the 27th ACM International Conference on Multimedia (MM ’19), October 21–25, 2019, Nice, France. ACM, New York, NY, USA, 4 pages. https://doi.org/10.1145/3343031.3350535.

Data Policy

The copyright of the dataset belongs to JAMSTEC. Please refer JAMSTEC’s data policy (https://www.jamstec.go.jp/e/database/data_policy.html) for details.

Conditions of Use

  • The dataset is licensed under CC BY-NC-SA 4.0 (https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en).
  • The dataset can be used for scientific or educational purposes only.
  • If you use the dataset, please cite it as follows:

    Takaki Nishio, Yuki Kawae (2025). J-EDI Organism Detection Dataset (JODD). JAMSTEC. https://doi.org/10.17596/0004035

Download

JediOrganismDetectionDataset.zip (343MB)

Contact

If you have any questions related to the dataset, please contact to:

jedi-admin@jamstec.go.jp