Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring
The number and areal extent of marine protected areas worldwide is rapidly increasing as a result of numerous national targets that aim to see up to 30% of their waters protected by 2030. Automated seabed classification algorithms are arising as faster and objective methods to generate benthic habit...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/10/1572 |
| _version_ | 1797567869191454720 |
|---|---|
| author | America Zelada Leon Veerle A.I. Huvenne Noëlie M.A. Benoist Matthew Ferguson Brian J. Bett Russell B. Wynn |
| author_facet | America Zelada Leon Veerle A.I. Huvenne Noëlie M.A. Benoist Matthew Ferguson Brian J. Bett Russell B. Wynn |
| author_sort | America Zelada Leon |
| collection | DOAJ |
| description | The number and areal extent of marine protected areas worldwide is rapidly increasing as a result of numerous national targets that aim to see up to 30% of their waters protected by 2030. Automated seabed classification algorithms are arising as faster and objective methods to generate benthic habitat maps to monitor these areas. However, no study has yet systematically compared their repeatability. Here we aim to address that problem by comparing the repeatability of maps derived from acoustic datasets collected on consecutive days using three automated seafloor classification algorithms: (1) Random Forest (RF), (2) K–Nearest Neighbour (KNN) and (3) K means (KMEANS). The most robust and repeatable approach is then used to evaluate the change in seafloor habitats between 2012 and 2015 within the Greater Haig Fras Marine Conservation Zone, Celtic Sea, UK. Our results demonstrate that only RF and KNN provide statistically repeatable maps, with 60.3% and 47.2% agreement between consecutive days. Additionally, this study suggests that in low-relief areas, bathymetric derivatives are non-essential input parameters, while backscatter textural features, in particular Grey Level Co-occurrence Matrices, are substantially more effective in the detection of different habitats. Habitat persistence in the test area between 2012 and 2015 was 48.8%, with swapping of habitats driving the changes in 38.2% of the area. Overall, this study highlights the importance of investigating the repeatability of automated seafloor classification methods before they can be fully used in the monitoring of benthic habitats. |
| first_indexed | 2024-03-10T19:48:30Z |
| format | Article |
| id | doaj.art-458823183fa5415babb43f7a9e691749 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T19:48:30Z |
| publishDate | 2020-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-458823183fa5415babb43f7a9e6917492023-11-20T00:34:45ZengMDPI AGRemote Sensing2072-42922020-05-011210157210.3390/rs12101572Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area MonitoringAmerica Zelada Leon0Veerle A.I. Huvenne1Noëlie M.A. Benoist2Matthew Ferguson3Brian J. Bett4Russell B. Wynn5University of Southampton, University Road, Southampton SO17 1BJ, UKNational Oceanography Centre, European Way, Southampton SO14 3ZH, UKNational Oceanography Centre, European Way, Southampton SO14 3ZH, UKJoint Nature Conservation Committee, Monkstone House, City Road, Peterborough PE1 1JY, UKNational Oceanography Centre, European Way, Southampton SO14 3ZH, UKWild New Forest CIC, 252 Woodlands Road, Woodlands, SO40 7GH, UKThe number and areal extent of marine protected areas worldwide is rapidly increasing as a result of numerous national targets that aim to see up to 30% of their waters protected by 2030. Automated seabed classification algorithms are arising as faster and objective methods to generate benthic habitat maps to monitor these areas. However, no study has yet systematically compared their repeatability. Here we aim to address that problem by comparing the repeatability of maps derived from acoustic datasets collected on consecutive days using three automated seafloor classification algorithms: (1) Random Forest (RF), (2) K–Nearest Neighbour (KNN) and (3) K means (KMEANS). The most robust and repeatable approach is then used to evaluate the change in seafloor habitats between 2012 and 2015 within the Greater Haig Fras Marine Conservation Zone, Celtic Sea, UK. Our results demonstrate that only RF and KNN provide statistically repeatable maps, with 60.3% and 47.2% agreement between consecutive days. Additionally, this study suggests that in low-relief areas, bathymetric derivatives are non-essential input parameters, while backscatter textural features, in particular Grey Level Co-occurrence Matrices, are substantially more effective in the detection of different habitats. Habitat persistence in the test area between 2012 and 2015 was 48.8%, with swapping of habitats driving the changes in 38.2% of the area. Overall, this study highlights the importance of investigating the repeatability of automated seafloor classification methods before they can be fully used in the monitoring of benthic habitats.https://www.mdpi.com/2072-4292/12/10/1572automated seafloor classificationmachine learning algorithmsbenthic habitat mapsautonomous underwater vehiclesGrey Level Co-occurrence Matricessidescan sonar |
| spellingShingle | America Zelada Leon Veerle A.I. Huvenne Noëlie M.A. Benoist Matthew Ferguson Brian J. Bett Russell B. Wynn Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring Remote Sensing automated seafloor classification machine learning algorithms benthic habitat maps autonomous underwater vehicles Grey Level Co-occurrence Matrices sidescan sonar |
| title | Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring |
| title_full | Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring |
| title_fullStr | Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring |
| title_full_unstemmed | Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring |
| title_short | Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring |
| title_sort | assessing the repeatability of automated seafloor classification algorithms with application in marine protected area monitoring |
| topic | automated seafloor classification machine learning algorithms benthic habitat maps autonomous underwater vehicles Grey Level Co-occurrence Matrices sidescan sonar |
| url | https://www.mdpi.com/2072-4292/12/10/1572 |
| work_keys_str_mv | AT americazeladaleon assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring AT veerleaihuvenne assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring AT noeliemabenoist assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring AT matthewferguson assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring AT brianjbett assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring AT russellbwynn assessingtherepeatabilityofautomatedseafloorclassificationalgorithmswithapplicationinmarineprotectedareamonitoring |