Search Help Show/Hide Menu
Summary
Item Identification
Keywords
Physical Location
Data Set Info
Support Roles
Extents
Access Info
Distribution Info
URLs
Tech Environment
Data Quality
Data Management
Lineage
Acquisition Info
Child Items
Related Items
Catalog Details

Summary

Short Citation
Pacific Islands Fisheries Science Center, 2024: Developing a semi-automated CoralNet Bleaching Classifier: annotations and imagery from survey sites across the Hawaiian Archipelago between 2014 and 2019, https://www.fisheries.noaa.gov/inport/item/67962.
Full Citation Examples

Abstract

This data package includes datasets and benthic imagery that support the development of a semi-automated coral bleaching classifier in CoralNet. The datasets in this package include annotations of imagery produced to test and train the CoralNet Bleaching Classifier, the annotation label set used in CoralNet, and site-level metadata for annotated imagery. The datasets were generated by scientists of the Ecosystem Sciences Division (ESD) at the Pacific Islands Fisheries Science Center (PIFSC) at NOAA, funded by the Coral Reef Conservation Program (CRCP). This data package includes non-ESD imagery used to develop this classifier, including images of sites in the Papahanoumokuakea National Monument (collected in 2014 by Courtney Couch, a scientist of the ESD), and of sites in Kaneohe Bay, Oahu (in 2015 by The Nature Conservancy). Other source imagery that was used to develop this classifier was collected by the ESD during the 2019 Hawaii coral bleaching event, as well as during the 2015, 2016, and 2019 NWHI National Coral Reef Monitoring Program (NCRMP) survey missions following the traditional photoquadrat collection methodology for NCRMP surveys; this imagery has already been archived, or is a pending archive package; they are described and accessible at https://www.fisheries.noaa.gov/inport/item/59193 and https://www.fisheries.noaa.gov/inport/item/36152.

To create this classifier, named the NOAA ESD Coral Bleaching Classifier (accessible at: https://coralnet.ucsd.edu/source/2947/), a set of training imagery with corresponding point coordinates and label annotations were uploaded to the classifier's source. The training annotations were made by human annotators on both randomly generated and targeted points. Once the training imagery and annotations were uploaded, the automated classification abilities of the classifier were tested by uploading imagery and point locations, but without corresponding label annotations. The machine generated labels for these points were then compared against the human generated labels, which had not been uploaded. See the CoralNet Bleaching Classifier SOP for more information: https://doi.org/10.25923/d0re-9y93.

Distribution Information

  • CSV - Comma Separated Values (Text)

    This file contains image level metadata for each image used to train and test the NOAA ESD Coral Bleaching classifier.

  • CSV - Comma Separated Values (Text)

    Labelset used to train and test the NOAA ESD Coral Bleaching Classifier; defines the short code annotations of the datasets.

  • CSV - Comma Separated Values (Text)

    This file contains point level comparison for the imagery used to test the NOAA ESD Coral Bleaching classifier. For every point, the machine suggested label and the human annotated label are shown.

  • CSV - Comma Separated Values (Text)

    This file contains the point-level labels assigned by human annotators to the imagery used to train the NOAA ESD Coral Bleaching Classifier.

  • JPEG, 7.82 MB

    Benthic images of one site off of Lisanski island and Midway Atoll in 2014, owned by Courtney Couch. This imagery was used to train the CoralNet Bleaching Classifier.

  • JPEG, 996 MB

    Benthic imagery owned by The Nature Conservancy of Kaneohe Bay in 2015. Imagery was used to train the CoralNet Bleaching Classifier, and also generated annotations for which data is accessible in this data package.

Access Constraints:

None

Use Constraints:

Please cite NOAA Fisheries, Ecosystem Science Division (ESD) when using the data.

Example:

Ecosystem Sciences Division; Pacific Islands Fisheries Science Center (2022). Developing a semi-automated CoralNet Bleaching Classifier: annotations and imagery from survey sites across the Hawaiian Archipelago between 2014 and 2019. NOAA's National Center for Environmental Information, https://www.fisheries.noaa.gov/inport/item/67962

Controlled Theme Keywords

biota, CORAL BLEACHING, DOC/NOAA/NMFS/PIFSC/ESD

Contact Information

Point of Contact
Courtney S Couch
courtney.s.couch@noaa.gov

Metadata Contact
Lori H Luers
lori.luers@noaa.gov

Extents

Geographic Area 1

-178.383718° W, -155.8366193° E, 28.459215° N, 19.31344° S

Main Hawaiian Islands (MHI), including Hawaii, Maui, Oahu, Molokai, and Lanai, and the Northwestern Hawaiian Islands (NWHI), including French Frigate Shoals, Lisianski Island, Pearl & Hermes Reef, Midway Atoll, and Kure Atoll.

Time Frame 1
2014-08-20 - 2019-11-14

Photoquadrat imagery used to train the CoralNet Bleaching Classifier was collected during multiple research missions from 2014 to 2019.

Item Identification

Title: Developing a semi-automated CoralNet Bleaching Classifier: annotations and imagery from survey sites across the Hawaiian Archipelago between 2014 and 2019
Short Name: CoralNet Bleaching Classifier
Status: Completed
Creation Date: 2022
Publication Date: 2022
Abstract:

This data package includes datasets and benthic imagery that support the development of a semi-automated coral bleaching classifier in CoralNet. The datasets in this package include annotations of imagery produced to test and train the CoralNet Bleaching Classifier, the annotation label set used in CoralNet, and site-level metadata for annotated imagery. The datasets were generated by scientists of the Ecosystem Sciences Division (ESD) at the Pacific Islands Fisheries Science Center (PIFSC) at NOAA, funded by the Coral Reef Conservation Program (CRCP). This data package includes non-ESD imagery used to develop this classifier, including images of sites in the Papahanoumokuakea National Monument (collected in 2014 by Courtney Couch, a scientist of the ESD), and of sites in Kaneohe Bay, Oahu (in 2015 by The Nature Conservancy). Other source imagery that was used to develop this classifier was collected by the ESD during the 2019 Hawaii coral bleaching event, as well as during the 2015, 2016, and 2019 NWHI National Coral Reef Monitoring Program (NCRMP) survey missions following the traditional photoquadrat collection methodology for NCRMP surveys; this imagery has already been archived, or is a pending archive package; they are described and accessible at https://www.fisheries.noaa.gov/inport/item/59193 and https://www.fisheries.noaa.gov/inport/item/36152.

To create this classifier, named the NOAA ESD Coral Bleaching Classifier (accessible at: https://coralnet.ucsd.edu/source/2947/), a set of training imagery with corresponding point coordinates and label annotations were uploaded to the classifier's source. The training annotations were made by human annotators on both randomly generated and targeted points. Once the training imagery and annotations were uploaded, the automated classification abilities of the classifier were tested by uploading imagery and point locations, but without corresponding label annotations. The machine generated labels for these points were then compared against the human generated labels, which had not been uploaded. See the CoralNet Bleaching Classifier SOP for more information: https://doi.org/10.25923/d0re-9y93.

Purpose:

This data package supports the detailed guidelines for developing a semi-automated classification tool in CoralNet to assist in the identification of coral bleaching.

Over the past decade, coral bleaching events have been increasing both in frequency and severity. In the Hawaiian Archipelago, three mass coral bleaching events have been recorded since 2014. With more frequent and severe global bleaching events on the horizon and the efficiency of collecting bleaching data using digital imagery, monitoring programs in Hawai'i need tools to efficiently extract bleaching data from images. Creating tools to quickly and accurately quantify the extent of coral bleaching is an important next step in developing strategies to appropriately respond to and monitor future mass bleaching events.

Other Citation Details:

The CoralNet Bleaching Classifier SOP will be publicly accessible at: https://doi.org/10.25923/d0re-9y93

Ehrenberg J, Winston M, Oliver T, Couch C. 2022. Development of a Semi-automated Coral Bleaching Classifier in CoralNet: A Summary of Standard Operating Procedures and Report of Results. U.S. Dept. of Commerce, NOAA Technical Memorandum NOAA-TM-NMFS-PIFSC-133, 11 p. https://doi.org/10.25923/d0re-9y93.

Supplemental Information:

Images used to create this coral bleaching classifier were collected in 2014, 2015, 2016, and 2019 during coral reef surveys across the Hawaiian Archipelago (including the Papahanoumokuakea Marine National Monument), by the Ecosystem Sciences Division (ESD) of the NOAA Pacific Islands Fisheries Science Center (PIFSC) and its partners with funding from NOAA's Coral Reef Conservation Program (CRCP), Project ID # 31361.

The imagery included in this data package are non-ESD owned and selected to train the tool based on benthic features.

Keywords

Theme Keywords

Thesaurus Keyword
Global Change Master Directory (GCMD) Data Center Keywords
DOC/NOAA/NMFS/PIFSC/ESD > Ecosystem Sciences Division, Pacific Islands Fisheries Science Center, National Marine Fisheries Service, NOAA, U.S. Department of Commerce
Global Change Master Directory (GCMD) Science Keywords
EARTH SCIENCE > OCEANS > COASTAL PROCESSES > CORAL REEFS > CORAL BLEACHING
ISO 19115 Topic Category
biota
UNCONTROLLED
CoRIS Discovery Thesaurus ONLY ONE CoRIS Discovery Thesaurus REQUIRED
CoRIS Theme Thesaurus EARTH SCIENCE > Biosphere > Zoology > Corals > Coral Diseases > Bleaching
CoRIS Theme Thesaurus EARTH SCIENCE > Biosphere > Zoology > Corals > Reef Monitoring and Assessment
CoRIS Theme Thesaurus EARTH SCIENCE > Oceans > Coastal Processes > Coral Reefs
CRCP Project 31361
CRCP Project Assessing impacts of Hawaii's 2019 coral bleaching event on coral recovery and coral communities
NODC DATA TYPES THESAURUS CORAL - BLEACHING
NODC DATA TYPES THESAURUS photograph
NODC INSTRUMENT TYPES THESAURUS camera
NODC OBSERVATION TYPES THESAURUS imagery
NODC OBSERVATION TYPES THESAURUS laboratory analyses
NODC PLATFORM NAMES THESAURUS Hi'ialakai
NODC PLATFORM NAMES THESAURUS small boat
NODC PROJECT NAMES THESAURUS Coral Reef Conservation Program
NODC PROJECT NAMES THESAURUS CORAL REEF STUDIES
NODC PROJECT NAMES THESAURUS Ocean Acidification Program
NODC SEA AREA NAMES THESAURUS: Central Pacific Ocean
NODC SEA AREA NAMES THESAURUS: Coastal Waters of Hawaii
NODC SEA AREA NAMES THESAURUS: North Pacific Ocean
NODC SUBMITTING INSTITUTION NAMES THESAURUS US DOC; NOAA; NMFS; Pacific Islands Fisheries Science Center; Ecosystem Sciences Division
None biota
None bleaching
None Coral Reef Ecosystem Division
None Coral Reef Ecosystem Program
None CoralNet
None CRED
None CREP
None Ecosystem Sciences Division
None ESD
None image analysis
None Pacific Islands Fisheries Science Center
None PIFSC
None Reef Assessment and Monitoring Program
None SOP

Spatial Keywords

Thesaurus Keyword
UNCONTROLLED
CoRIS Place Thesaurus COUNTRY/TERRITORY > United States of America > Hawaiian Islands (21N157W0027)
CoRIS Place Thesaurus OCEAN BASIN > Pacific Ocean > Central Pacific Ocean > Hawaiian Islands (21N157W0027)

Instrument Keywords

Thesaurus Keyword
Global Change Master Directory (GCMD) Instrument Keywords
Computer > Computer
UNCONTROLLED
NODC INSTRUMENT TYPES THESAURUS computer

Physical Location

Organization: Pacific Islands Fisheries Science Center
City: Honolulu
State/Province: HI
Country: USA

Data Set Information

Data Set Scope Code: Data Set
Maintenance Frequency: None Planned
Entity Attribute Overview:

This data package includes:

Labelset used to train classifier (and defines annotations in datasets):

1) NOAA_ESD_CoralBleachingClassifier_Labelset

Annotations for the Training Sites:

2) NOAA_ESD_CoralBleachingClassifier_TrainingAnnotations

Human/Machine Annotations for Test Sites (results):

3) NOAA_ESD_CoralBleachingClassifier_TestAnnotations

Image Metadata:

4) NOAA_ESD_CoralBleachingClassifier_ImageMetadata

Imagery:

5) 2015 imagery owned by The Nature Conservancy

6) 2014 imagery owned from Courtney Couch from the Northwestern Hawaiian Islands, Hawaii Island and Kaneohe Bay (Oahu), archived with NCEI; used as training source

NOAA ESD Coral Bleaching Classifier Annotations Data Dictionary defines columns of annotation datasets and can be accessed at:

https://www.fisheries.noaa.gov/inport/item/68138

NOAA ESD Coral Bleaching Classifier Image Metadata Data Dictionary defines the image metadata file and can be accessed at:

https://www.fisheries.noaa.gov/inport/item/68141

Distribution Liability:

While every effort has been made to ensure that these data are accurate and reliable within the limits of the current state of the art, NOAA cannot assume liability for any damages caused by errors or omissions in the data, nor as a result of the failure of the data to function on a particular system. NOAA makes no warranty, expressed or implied, nor does the fact of distribution constitute such a warranty.

Data Set Credit: NOAA Fisheries, ESD and funded by the NOAA CRCP

Support Roles

Data Set Credit

CC ID: 1186036
Date Effective From: 2022
Date Effective To:
Contact (Organization): NOAA Coral Reef Conservation Program (CRCP)
Address: 1305 East West Highway 10th Floor
Silver Spring, MD 20910-3281
Phone: (301) 713-3155
URL: https://coralreef.noaa.gov

Data Steward

CC ID: 1186038
Date Effective From: 2022
Date Effective To:
Contact (Person): Winston, Morgan S
Address: 1845 Wasp Blvd.
Honolulu, HI 96818
USA
Email Address: morgan.winston@noaa.gov
Phone: (808)725-5480

Distributor

CC ID: 1186039
Date Effective From: 2022
Date Effective To:
Contact (Organization): National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD)
Address: NOAA/NESDIS E/OC SSMC3, 4th Floor, 1351 East-West Highway
Silver Spring, MD 20910-3282
Phone: (301) 713-3277

Metadata Contact

CC ID: 1186040
Date Effective From: 2022
Date Effective To:
Contact (Person): Luers, Lori H
Address: 1845 Wasp Blvd.
Honolulu, HI 96818
USA
Email Address: lori.luers@noaa.gov

Originator

CC ID: 1186035
Date Effective From: 2022
Date Effective To:
Contact (Organization): Pacific Islands Fisheries Science Center (PIFSC)
Address: 1845 Wasp Blvd.
Honolulu, HI 96818
USA
Email Address: pifsc.info@noaa.gov
Phone: 808-725-5360
URL: https://www.pifsc.noaa.gov
Business Hours: 8:00 a.m. - 4:30 p.m.

Point of Contact

CC ID: 1186037
Date Effective From: 2022
Date Effective To:
Contact (Person): Couch, Courtney S
Email Address: courtney.s.couch@noaa.gov

Extents

Currentness Reference: Ground Condition

Extent Group 1

Extent Group 1 / Geographic Area 1

CC ID: 1188450
W° Bound: -178.383718
E° Bound: -155.8366193
N° Bound: 28.459215
S° Bound: 19.31344
Description

Main Hawaiian Islands (MHI), including Hawaii, Maui, Oahu, Molokai, and Lanai, and the Northwestern Hawaiian Islands (NWHI), including French Frigate Shoals, Lisianski Island, Pearl & Hermes Reef, Midway Atoll, and Kure Atoll.

Extent Group 1 / Time Frame 1

CC ID: 1188449
Time Frame Type: Range
Start: 2014-08-20
End: 2019-11-14
Description:

Photoquadrat imagery used to train the CoralNet Bleaching Classifier was collected during multiple research missions from 2014 to 2019.

Access Information

Security Class: Unclassified
Security Classification System:

Not applicable

Security Handling Description:

Not applicable

Data Access Policy:

NOAA Ecosystem Science Division (ESD) Data Sharing Recommendations, version 9.0 updated August 12, 2015:

ESD welcomes the opportunity to collaborate on research issues contributing to the scientific basis for better management of marine ecosystems. ESD has a very diverse set of field activities that generates large volumes of data using an array of data collection protocols.

The following recommendations are for your consideration as you use this data:

1) Data analyses should take all field exigencies into account. The most effective way to do this would be active collaboration with ESD principal investigators.

2) In all presentations, product releases, or publications using data generated by ESD, proper acknowledgement of both ESD and the individuals responsible for data collection is expected. Citing the DOI (if available) is preferred, a non-DOI example is listed below.

3) If you collect or generate data for the same study areas, ESD requests that you share relevant information on complimentary data collections.

4) Those receiving data are strongly urged to inform the ESD Data Management Team of any errors and discrepancies that are discovered during the course of using these data. They are further urged to bring to the attention of the Team all problems and difficulties encountered in using these data. This information is necessary in order to improve the collections and to facilitate more efficient and economical data processing and retrieval. The users are asked to supply copies of any missing data that may be located, and to provide information as to significant subsets and special aggregations of data that are developed in using the material provided.

Example citation:

"This publication makes use of data products provided by the Ecosystem Science Division (ESD), Pacific Islands Fisheries Science Center (PIFSC), National Marine Fisheries Service (NMFS), National Oceanic and Atmospheric Administration (NOAA), with funding support from the NOAA Coral Reef Conservation Program (CRCP). The analysis and interpretations presented here are solely that of the current authors."

Data Access Procedure:

Data can be accessed online via the NOAA National Centers for Environmental Information (NCEI) Ocean Archive.

Data Access Constraints:

None

Data Use Constraints:

Please cite NOAA Fisheries, Ecosystem Science Division (ESD) when using the data.

Example:

Ecosystem Sciences Division; Pacific Islands Fisheries Science Center (2022). Developing a semi-automated CoralNet Bleaching Classifier: annotations and imagery from survey sites across the Hawaiian Archipelago between 2014 and 2019. NOAA's National Center for Environmental Information, https://www.fisheries.noaa.gov/inport/item/67962

Metadata Access Constraints:

None

Metadata Use Constraints:

None

Distribution Information

Distribution 1

CC ID: 1186041
Start Date: 2022-10-27
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: NOAA_ESD_CoralBleachingClassifier_ImageMetadata.csv
Description:

This file contains image level metadata for each image used to train and test the NOAA ESD Coral Bleaching classifier.

File Type (Deprecated): .csv
Distribution Format: CSV - Comma Separated Values (Text)

Distribution 2

CC ID: 1188997
Start Date: 2022-10-27
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: NOAA_ESD_CoralBleachingClassifier_Labelset.csv
Description:

Labelset used to train and test the NOAA ESD Coral Bleaching Classifier; defines the short code annotations of the datasets.

File Type (Deprecated): csv
Distribution Format: CSV - Comma Separated Values (Text)

Distribution 3

CC ID: 1186355
Start Date: 2022-10-27
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: NOAA_ESD_CoralBleachingClassifier_TestAnnotations.csv
Description:

This file contains point level comparison for the imagery used to test the NOAA ESD Coral Bleaching classifier. For every point, the machine suggested label and the human annotated label are shown.

File Type (Deprecated): .csv
Distribution Format: CSV - Comma Separated Values (Text)

Distribution 4

CC ID: 1186352
Start Date: 2022-10-27
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: NOAA_ESD_CoralBleachingClassifier_TrainingAnnotations.csv
Description:

This file contains the point-level labels assigned by human annotators to the imagery used to train the NOAA ESD Coral Bleaching Classifier.

File Type (Deprecated): .csv
Distribution Format: CSV - Comma Separated Values (Text)

Distribution 5

CC ID: 1190881
Start Date: 2022
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: Couch_Training_Images_NWHI_2014
Description:

Benthic images of one site off of Lisanski island and Midway Atoll in 2014, owned by Courtney Couch. This imagery was used to train the CoralNet Bleaching Classifier.

File Type (Deprecated): JPG
Distribution Format: JPEG
File Size: 7.82 MB

Distribution 6

CC ID: 1190880
Start Date: 2022
End Date: Present
Download URL: https://www.ncei.noaa.gov/archive/accession/0269246
Distributor: National Centers for Environmental Information - Silver Spring, Maryland (NCEI-MD) (2022 - Present)
File Name: TNC_Images_Kaneohe_2015
Description:

Benthic imagery owned by The Nature Conservancy of Kaneohe Bay in 2015. Imagery was used to train the CoralNet Bleaching Classifier, and also generated annotations for which data is accessible in this data package.

File Type (Deprecated): JPG
Distribution Format: JPEG
File Size: 996 MB

URLs

URL 1

CC ID: 1186042
URL: https://www.fisheries.noaa.gov/region/pacific-islands#science
Name: NOAA Fisheries Pacific Islands Region - Science Overview
URL Type:
Online Resource

URL 2

CC ID: 1186358
URL: https://coralnet.ucsd.edu/pages/help/api/
Name: CoralNet API Documentation
URL Type:
Online Resource
File Resource Format: html
Description:

The CoralNet deploy API lets you programmatically classify any image using any automatic classifier on CoralNet. You specify images using web URLs along with the row, column locations you want classified and you get the automated classifications back. Tutorial & Reference links provided.

URL 3

CC ID: 1186359
URL: https://coralnet.ucsd.edu/source/2947/
Name: NOAA ESD CoralNet Bleaching Classifier
Description:

The trained bleaching classifier is publicly available (Global ID 24223). If your organization is interested in uploading photoquadrat imagery to derive percent bleaching cover estimates, please contact Morgan Winston (morgan.winston@noaa.gov)

URL 4

CC ID: 1190883
URL: https://doi.org/10.25923/d0re-9y93
Name: Development of a Semi-automated Coral Bleaching Classifier in CoralNet: A Summary of Standard Operating Procedures and Report of Results
URL Type:
Online Resource
Description:

In this report, we provide a description of how to create a coral bleaching classifier in CoralNet using photoquadrat imagery. We also include a summary of accuracy of NOAA ESD's classifier at both the point and site level. The trained bleaching classifier is publically available (Global ID 24223) and can be accessed using the CoralNet Application Program Interface (API).

URL 5

CC ID: 1190884
URL: https://github.com/morgan-winston/coralnet-bleaching-classifier-development
Name: Coral Bleaching R Scripts
URL Type:
Online Resource
Description:

This repository contains code used to develop a CoralNet classifier to automate annotations of coral bleaching: "Point_Annotator_Raster_v1.3.R" and specifically for the CoralNet Bleaching Classiifer "V3__Bleaching_Site_Sorting.R"

URL 6

CC ID: 1198574
URL: https://www.fisheries.noaa.gov/resource/document/development-semi-automated-coral-bleaching-classifier-coralnet-summary-standard
Name: PIFSC Resource Report
URL Type:
Online Resource
Description:

PIFSC landing page for Coral Bleaching Classifier Report

Technical Environment

Description:

Rstudio for working with annotations and web-based CoralNet annotation tool.

Data Quality

Accuracy:

Good image quality is essential for using this classifier, also referred to as source in CoralNet, for developing a new classifier . Poor image quality (pixelated, discolored, or unfocused) will make it difficult for annotators/the classifier to correctly label annotations. Annotators should be well trained and calibrated to ensure the source's classifier is being trained correctly. Lastly, labels need to be clearly defined to reduce confusion for annotators.

Completeness Report:

Only shallow water (0 to 30 meter depth range) hard-bottom habitats were surveyed.

Conceptual Consistency:

The same labelset was used throughout the training of the classifier.

Quality Control Procedures Employed:

Quality control is enforced by means of point-to-point, inter-observer calibration exercises that are conducted before each image analysis production series. Training modules and standard operating procedures have also been developed and documented to ensure improved performance and consistent imagery analysis results produced by multiple analysts.

Data Management

Have Resources for Management of these Data Been Identified?: Yes
Approximate Percentage of Budget for these Data Devoted to Data Management: Unknown
Do these Data Comply with the Data Access Directive?: Yes
Is Access to the Data Limited Based on an Approved Waiver?: No
Approximate Delay Between Data Collection and Dissemination: Unknown
Actual or Planned Long-Term Data Archive Location: NCEI-MD
Approximate Delay Between Data Collection and Archiving: Unknown
How Will the Data Be Protected from Accidental or Malicious Modification or Deletion Prior to Receipt by the Archive?:

NOAA IRC and NOAA Fisheries ITS resources and assets.

Lineage

Lineage Statement:

To create this classifier, named the NOAA ESD Coral Bleaching Classifier (see 'URLs' in this InPort record for a link to the CoralNet source), a set of training imagery with corresponding point coordinates and label annotations were uploaded to the classifier's source. The training annotations (hard coral, bleached hard coral, and unknown) were made by human annotators on both randomly generated and targeted points. Once the training imagery and annotations were uploaded, the automated classification abilities of the classifier were tested by uploading imagery and point locations, but without corresponding label annotations. The machine generated labels for these points were then compared against the human generated labels, which had not been uploaded. See the CoralNet Bleaching Classifier SOP for more information (link included under 'URLs' in this InPort record).

Sources

Ehrenberg J., M. Winston, T. Oliver, and C. Couch (2022). Development of a coral bleaching classifier tool for CoralNet: a summary of standard operating procedures and report of results. Pacific Islands Fish. Sci. Cent., Natl. Mar. Fish. Serv., NOAA, Honolulu, HI 96818-5007. Pacific Islands Fish. Sci. Cent. Admin. Rep. H-17-YY, xx p.

CC ID: 1191439
Contact Role Type: Originator
Contact Type: Organization
Contact Name: Pacific Islands Fisheries Science Center
Citation URL: https://doi.org/10.25923/d0re-9y93
Citation URL Name: doi
Citation URL Description:

In this report, we provide a description of how to create a coral bleaching classifier in CoralNet using photoquadrat imagery. We also include a summary of accuracy of NOAA ESD's classifier at both the point and site level. The trained bleaching classifier is publically available (Global ID 24223) and can be accessed using the CoralNet Application Program Interface (API).

Source Contribution:

Details the SOP and data associated with developing the CoralNet Bleaching Classifier tool.

Process Steps

Process Step 1

CC ID: 1186306
Description:

To train a bleaching classifier using CoralNet, the ESD created a new source (CREP-HAWAII BLEACHING v2: Classifier ID 23472) (steps 1-4) containing imagery collected across 178 sites in the Hawaiian Archipelago during the 2019 mass coral bleaching event. A total of 5207 images were used, with approximately 30 images for each site; however, image numbers ranged from 5 - 72 images per site. The imagery was manually annotated using the simple labelset. Once the initial round manual of annotations were completed, a subset of sites with the highest prevalence of coral bleaching were selected and targeted annotations were added manually. CREP-HAWAII BLEACHING v2 served as an experimental test source to confirm that annotations created using steps 5-40 could be uploaded to CoralNet and be used to train a new source.

Process Contact: Winston, Morgan S
Phone (Voice): (808)725-5480
Email Address: morgan.winston@noaa.gov

Process Step 2

CC ID: 1186307
Description:

To test the machine accuracy of the bleaching classifier, 80% of the sites were designated as training sites and uploaded to the NOAA ESD Coral Bleaching Classifier source using the default feature extractor EfficentNet, while the remaining 20% served as test sites for the source. Sites were randomly sorted. There were a total of 143 training sites selected for NOAA ESD Coral Bleaching Classifier; 129 sites had only random annotations and 14 sites had targeted and random annotations. The resulting 48,622 training annotations were uploaded to NOAA ESD Coral Bleaching Classifier and automated machine classification accuracy was obtained.

Process Contact: Winston, Morgan S
Phone (Voice): (808)725-5480
Email Address: morgan.winston@noaa.gov

Process Step 3

CC ID: 1186308
Description:

A total of 35 sites were randomly selected as test sites (sites with only random annotations). Test site imagery was uploaded without annotation labels to the NOAA ESM Coral Bleaching Classifier source. Machine suggestions were downloaded and compared to the original human annotations. From this comparison, machine accuracy was calculated as the proportion of machine points that matched with the human annotations.

Process Contact: Winston, Morgan S
Phone (Voice): (808)725-5480
Email Address: morgan.winston@noaa.gov

Acquisition Information

Instruments

Instrument Unavailable Reason: Not Applicable

Platforms

Platform Unavailable Reason: Not Applicable

Related Items

Item Type Relationship Type Title
Data Set (DS) Series Benthic Cover Derived from Analysis of Benthic Images Collected in the Hawaiian Archipelago During the 2019 Bleaching Event

NCRMP benthic cover survey by which images are annotated to produce benthic cover estimates of communities. This analysis was unique in that it adapted the CoralNet label set to allow for identification of bleached corals along with the typical benthic cover estimates.

Data Set (DS) Cross Reference Benthic Images from Stratified Random Site (StRS) Surveys in the Main Hawaiian Islands During the 2019 Bleaching Event

Includes ESD-owned source imagery which was annotated to produce datasets part of the CoralNet Bleaching Classifier data package.

Data Set (DS) Cross Reference National Coral Reef Monitoring Program: Benthic Images Collected from Stratified Random Sites (StRS) across the Hawaiian Archipelago from 2013 to 2019

Includes ESD-owned source imagery which was annotated to produce datasets part of the CoralNet Bleaching Classifier data package.

Catalog Details

Catalog Item ID: 67962
GUID: gov.noaa.nmfs.inport:67962
Metadata Record Created By: Lori H Luers
Metadata Record Created: 2022-10-01 02:20+0000
Metadata Record Last Modified By: SysAdmin InPortAdmin
Metadata Record Last Modified: 2023-10-17 16:12+0000
Metadata Record Published: 2022-11-10
Owner Org: PIFSC
Metadata Publication Status: Published Externally
Do Not Publish?: N
Metadata Last Review Date: 2022-11-04
Metadata Review Frequency: 1 Year
Metadata Next Review Date: 2023-11-04