Research Data

DeepFish: A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis

Details

Print this Page Print this Page
Show all sections
General
Title
DeepFish: A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis
Type
Dataset
Date Record Created
2020-09-15
Date Record Modified
2020-09-16
Language
English
Coverage
Date Coverage
(no information)
Time Period
(no information)
Geospatial Location
  • Palm Islands, Queensland, Australia
  • Western Australia
Description
Descriptions
  1. Type: full

    The dataset consists of approximately 40 thousand images collected underwater from 20 habitats in the marine-environments of tropical Australia.

    The dataset originally contained only classification labels. Thus, we collected point-level and segmentation labels to have a more comprehensive fish analysis benchmark.

    Videos for DeepFish were collected for 20 habitats from remote coastal marine environments of tropical Australia. These videos were acquired using cameras mounted on metal frames, deployed over the side of a vessel to acquire video footage underwater. The cameras were lowered to the seabed and left to record the natural fish community, while the vessel maintained a distance of 100 m. The depth and the map coordinates of the cameras were collected using an acoustic depth sounder and a GPS, respectively. Video recording was carried out during daylight hours and in relatively low turbidity periods. The video clips were captured in full HD resolution (1920 × 1080 pixels) from a digital camera. In total, the number of video frames taken is 39,766. 

Related Publications
(no information)
Related Websites
(no information)
Related Data
(no information)
Related Services
(no information)
Technical metadata
(no information)
People
Creators
  1. Managed by: Mr Alzayat Saleh , alzayat.saleh@jcu.edu.au , College of Science & Engineering, ARC Centre of Excellence for Coral Reef Studies
  2. Owned by: Dr Michael Bradley , michael.bradley@jcu.edu.au , Marine Biology & Aquaculture, Zoology and Ecology
  3. Associated with: Prof Marcus Sheaves , marcus.sheaves@jcu.edu.au , Division of Global Strategy & Engagement, Zoology and Ecology, Centre for Tropical Water and Aquatic Ecosystem Research
Primary Contact
Mr Alzayat Saleh, alzayat.saleh@jcu.edu.au
Supervisors
  1. Prof Marcus Sheaves , marcus.sheaves@jcu.edu.au
Collaborators
  1. Issam H. Laradji, Element AI, Montreal, Canada & University of British Columbia, Vancouver, Canada
  2. David Vazquez, Element AI, Montreal, Canada
  3. Dmitry A. Konovalov, James Cook University
Subject
Fields of Research
  1. 080104 - Computer Vision (080104)
Socio-Economic Objective
  1. 830199 - Fisheries - Aquaculture not elsewhere classified (830199)
Keywords
  1. fish datasets
  2. DeepFish
  3. deep learning
  4. Computer Vision
  5. fish Segmentation
  6. ARC Centre of Excellence for Coral Reef Studies
Research Activity
(no information)
Research Themes
Tropical Ecosystems, Conservation and Climate Change
Industries and Economies in the Tropics
Tropical Health, Medicine and Biosecurity
Rights
License
CC BY 4.0: Attribution 4.0 International
License - Other
(no information)
Access Rights/Conditions
Open access. If the data is not freely accessible via the link provided, please contact the nominated data manager or researchdata@jcu.edu.au for assistance.
Type
open
Rights
(no information)
Data
Data Location
Online Locations
  1. https://cloudstor.aarnet.edu.au/plus/s/NfjObIhtUYO6332
Stored At
(no information)
Citation
Cite:
Saleh, A.; Bradley, M.; Sheaves, M. (2020): DeepFish: A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis. James Cook University. (dataset). http://dx.doi.org/10.25903/5f617fb6d6e0e
Digital Object Identifier (DOI):
10.25903/5f617fb6d6e0e