Dataset: SPC-UW-ZooCam classified zooplankton images

Name: Classified Zooplankton ZooCam Images Captured by the Hoodsport ORCA Profiling Mooring Mounted SPC-2 Zoocam in the Hood Canal, Puget Sound, Washington from July to September 2018 (Zooplankton Swimming project)
Published: 2024-05-14
Keywords: zooplankton, in-situ images, camera, Puget Sound, oceans

View Data: Data not available yet

Cite This Dataset

Spatial Extent: N:47.421817 E:-123.112583 S:47.421817 W:-123.112583

Hood Canal, Puget Sound, Washington, USA

Temporal Extent: 2018-07-09 - 2018-09-22

Project:

Causes and consequences of hypoxia and pH impacts on zooplankton: Linking movement behavior to vertical distribution. (Zooplankton Swimming)

Principal Investigator:

Julie E. Keister (University of Washington, UW)

Co-Principal Investigator:

Daniel Grunbaum (University of Washington, UW)

Student:

Deana Crouser (National Oceanic and Atmospheric Administration - Alaska Fisheries Science Center, NOAA-AFSC)

Amy Wyeth (University of Washington, UW)

Contact:

Julie E. Keister (University of Washington, UW)

BCO-DMO Data Manager:

Sawyer Newman (Woods Hole Oceanographic Institution, WHOI BCO-DMO)

Version:

Version Date:

2024-05-14

Restricted:

Validated:

Current State:

Preliminary and in progress

Classified Zooplankton ZooCam Images Captured by the Hoodsport ORCA Profiling Mooring Mounted SPC-2 Zoocam in the Hood Canal, Puget Sound, Washington from July to September 2018 (Zooplankton Swimming project)

Abstract:

This dataset consists of images of individual zooplankton taken by an in-situ camera system (the SPC UW ZooCam) that was deployed on the Hoodsport ORCA profiling mooring in Hood Canal (Puget Sound), WA in summer 2018. Images were taxonomically identified by expert zooplankton ecologists. These images are sorted into folders by taxonomic identification and were used as a training set for Machine Learning classification of unknown images to study the behavior of zooplankton under varying ocean conditions.

Expand/Collapse All

Description

Image Directory Category Descriptions

ClusteredSnow - Non-symmetrical "aggregated" clusters of marine snow
Filament_Filaments - singular and/or multiple, slender filament/threadlike objects
MarineSnow - Non-symmetrical "scattered" snow (smaller and numerous vs ClusteredSnow)
Unknown - Images that are too blurry to confidently classify or are "unknown"

Gelatinous:

Anthomedusae - (Euphysa tentaculata)
Cydippida - (Euplokamis dunlapae)
Cydippida - (Pleurobrachia bachei)
Cydippida - (Unknown)

Lobata:

Siphonophore - Calycophorae (Muggiaea atlantica)
Trachymedusae - (Aglantha digitale)
Trachymedusae - (Pantachogon haeckeli)
Trachymedusae - (young)

Copepoda:

Calanoida - (Acartia spp.)
Calanoida - (Calanus sp.)
Calanoida - (Centropages abdominalis)
Calanoida - (Metridia spp.)
Calanoida - (Psudo Micro Para) - includes pseudocalanus, microcalanus, and paracalanus
Copepoda - (nauplii)
Cyclopoida - (Oithona spp.)
Harpacticoida - (Microsetella rosea)
Poecilostomatoida - (Ditrichocoryceaus anglicus)
Poecilostomatoida - (Triconia spp.)

Amphipoda:

Cyphocariadae - (Cyphocaris challengeri)
Hyperiidea - (Themisto pacifica_Hyperoche sp.)
Gammeridea - (possibly Calliopius sp.)

Phytoplankton:

Dinoflagellata - (Noctiluca)
(Diatoms)

Other:

Chaetognatha
Decapoda - Caridea (Shrimp)
Euphausiacea - Euphausiidae (Krill)
Ostracoda - (Halocyprididae)
Fish_larvae
Larvacea - (Oikopleaura dioica)
Pteropoda - (Clione limacina)
Ptreropoda - (Limacina helicina)

Methods & Sampling

These images were collected by the SPC-UW-Zoocam, which is an underwater imaging system that was mounted on a profiling mooring. Images were collected during profiling of the mooring at depths between 5 - 110 meters.

Image data processing (Region of Interest (ROI) selection) was conducted internally, with processed and compressed images streamed to shore over cellular networks.

Problems/Issues

Many images are blurry or the organism was imaged at a non-oblique angle, making taxonomic classifications somewhat subjective, even for trained taxonomists. The classifications represent the best possible visual identification, but should not be considered definitive. Images that are too blurry to confidently classify are stored within the "Unknown" folder in the zipped image directory file.

Data Processing Description

Image processing

Images were visually (manually) identified for taxonomy and sorted into folders. In this dataset, some images were duplicated and rotated to create an augmented dataset to train a Machine Learning algorithm (labeled as “augmented” in each file name).

Images that are too blurry to confidently classify are stored within the "Unknown" folder in the zipped image directory file.

Image File Naming Convention

File name example: SPC-UW-1534792542343244-337276376507-004392-154-2546-848-756-780.png

SPC: Scripps Plankton Camera

UW: University of Washington

1534792542343244: Unix Date/Timestamp [To convert to UTC, use (X / 86400) + 25569.]

337276376507: Directory and subdirectory extension specific to this directory organization system.

154-2546-848-756-780: ROI (Region of Interest) details (x, y, w, h, image size)

Processing for Reuse

This dataset needs to undergo pre-processing steps (i.e., image augmentation and balancing) before it should be used to train an algorithm for best results.

BCO-DMO Processing

- Latitude and longitude values added to the data file
- augmented_image_flag column added to indicate if an image file has gone through augmentation
- A human readable datetime column was created from the unix_datetimestamp value present in the filename
- roi_x, roi_y, roi_h, and roi_image_size columns added from the image file name; this was parsed from the image file name

More Information about this dataset

Funding Sources

Funding Source	Award Number
NSF Division of Ocean Sciences	OCE-1657992

Deployments

Deployment	Synonyms	Start Date	Platform	Investigator
UW_SPC-2_Zoocam		2018-06-26	ORCA-UW-Hoodsport	Julie E. Keister (Chief Scientist)	data info

Instruments

Underwater Camera

Supplied Name: SPC-UW-Zoocam

Supplied Description:

The SPC-UW-Zoocam designed and built specifically for this project. The Zoocam was custom-built by Paul Roberts in the Jaffe Imaging Laboratory at the University of California San Diego. It is an underwater camera system that captured still images of zooplankton within a 500-mL imaged area using lighting in the visible wavelength range.

Instrument Type

Generic Name: Underwater Camera

Community Identifier: http://vocab.nerc.ac.uk/collection/L05/current/180/

Generic Description:

All types of photographic equipment that may be deployed underwater including stills, video, film and digital systems.

Parameters

Date and time formats are described in python strftime() syntax.

Supplied Name	Supplied description	Supplied Units	Standard Name
filename	Filename of an individual zooplankton image, i.e., region of interest (ROI)	unitless	file_name
relative_filepath	Relative directory path where the "filename" image is stored within the zip file, SPC-UW-ZooCam_Training_Photos.zip	unitless	file_name
filesize_bytes	Image file size	bytes	file_size
md5sum	Image file md5sum value	unitless	no_bcodmo_term
latitude	Latitude of mooring location where the ZooCam camera was deployed and the image file was collected in decimal degrees; a positive value indicates a Northern coordinate	decimal degrees	lat
longitude	Longitude of mooring location where the ZooCam camera was deployed and the image file was collected in decimal degrees; a negative value indicates a Western coordinate	decimal degrees	lon
augmented_image_flag	A "augmented_image" value indicates that an image was duplicated and rotated to create an augmented dataset; blank data values in this column indicate that an image was not augmented	unitless	flag
datetime_utc	Datetime value derived from the unix timestamp represented in the image filename. Format: %Y-%m-%d %H:%M:%S	unitless	DateTime
roi_x	ROI (region of interest) x value of image file	image_analysis	no_bcodmo_term
roi_y	ROI (region of interest) y value of image file	image_analysis	no_bcodmo_term
roi_w	ROI (region of interest) w value of image file	image_analysis	no_bcodmo_term
roi_h	ROI (region of interest) h value of image file	image_analysis	no_bcodmo_term
roi_image_size	ROI (region of interest) image size	pixels	no_bcodmo_term

Status message

Database

Contribute Data

Dataset: SPC-UW-ZooCam classified zooplankton images