Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking
Coral reefs are vital for biodiversity, coastal protection, food security, and tourism, yet they face severe threats from anthropogenic activities and climate change, which are leading to their decline. Effective coral reef monitoring is essential for ecological understanding and conservation, but t...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-11-01
|
| Series: | Ecological Informatics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1574954125001797 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850235360701841408 |
|---|---|
| author | Younes Ouassine Noël Conruyt Mohsen Kayal Philippe A. Martin Lionel Bigot Vignes Lebbe Regine Hajar Moussanif Jihad Zahir |
| author_facet | Younes Ouassine Noël Conruyt Mohsen Kayal Philippe A. Martin Lionel Bigot Vignes Lebbe Regine Hajar Moussanif Jihad Zahir |
| author_sort | Younes Ouassine |
| collection | DOAJ |
| description | Coral reefs are vital for biodiversity, coastal protection, food security, and tourism, yet they face severe threats from anthropogenic activities and climate change, which are leading to their decline. Effective coral reef monitoring is essential for ecological understanding and conservation, but traditional methods are resource-intensive and rely on experts. To address these challenges, we present an automated, deep learning-based monitoring system that integrates YOLOv8, a state-of-the-art object detection algorithm, with DeepSORT, a robust multi-object tracking method, to identify and track coral formations in underwater video footage. Our system was fine-tuned using two curated and annotated datasets: AIMECORAL1 (580 images from the Southwest Indian Ocean) and AIMECORAL2 (282 images from New Caledonia, Pacific Ocean), encompassing diverse coral species and environmental conditions. The system's performance was evaluated using established metrics: object detection precision, Multiple Object Tracking Accuracy (MOTA), Multiple Object Tracking Precision (MOTP), and Identity F1 Score (IDF1). Precision improved from 59.9 % (after fine-tuning on AIMECORAL1) to 84.7 % on the combined datasets. The tracking system achieved a MOTA of 82.63 %, MOTP of 83.28 %, and IDF1 of 70.76 %, demonstrating reliable multi-object tracking in complex underwater environments. We applied our framework to a case study involving video transects from an outer reef site in New Caledonia, comparing data from 2021 and 2022. This automated solution offers a scalable, cost-effective alternative to traditional monitoring methods, supporting seamless, large-scale reef assessment. By leveraging deep learning, our approach enables more efficient data collection, contributing to the protection of these vulnerable ecosystems in the face of increasing environmental pressures. |
| format | Article |
| id | doaj-art-80a0cb824fe14065b3bed8a3c3bf46bc |
| institution | OA Journals |
| issn | 1574-9541 |
| language | English |
| publishDate | 2025-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Ecological Informatics |
| spelling | doaj-art-80a0cb824fe14065b3bed8a3c3bf46bc2025-08-20T02:02:17ZengElsevierEcological Informatics1574-95412025-11-018910317010.1016/j.ecoinf.2025.103170Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT trackingYounes Ouassine0Noël Conruyt1Mohsen Kayal2Philippe A. Martin3Lionel Bigot4Vignes Lebbe Regine5Hajar Moussanif6Jihad Zahir7Computer Science Department, LISI, Cadi Ayyad University, Morocco; Corresponding author.EA2525 LIM, (I.T. Department), University of La Réunion, 97400 Saint-Denis, FranceENTROPIE, IRD, IFREMER, CNRS, University of La Reunion, University of New Caledonia, Noumea, New Caledonia; Station Marine d'Endoume, Marseille, FranceEA2525 LIM, (I.T. Department), University of La Réunion, 97400 Saint-Denis, FranceEA2525 LIM, (I.T. Department), University of La Réunion, 97400 Saint-Denis, FranceInstitut de Systématique, Evolution, Biodiversité (ISYEB), SU, MNHN, CNRS, EPHE, UA - CP 48, 57 rue Cuvier, 75005 Paris, FranceComputer Science Department, LISI, Cadi Ayyad University, MoroccoComputer Science Department, LISI, Cadi Ayyad University, Morocco; EA2525 LIM, (I.T. Department), University of La Réunion, 97400 Saint-Denis, France; ENTROPIE, IRD, IFREMER, CNRS, University of La Reunion, University of New Caledonia, Noumea, New Caledonia; Institut de Systématique, Evolution, Biodiversité (ISYEB), SU, MNHN, CNRS, EPHE, UA - CP 48, 57 rue Cuvier, 75005 Paris, France; Station Marine d'Endoume, Marseille, France; UMMISCO, IRD, Computer Science Department, LISI, CADI Ayyad University, MoroccoCoral reefs are vital for biodiversity, coastal protection, food security, and tourism, yet they face severe threats from anthropogenic activities and climate change, which are leading to their decline. Effective coral reef monitoring is essential for ecological understanding and conservation, but traditional methods are resource-intensive and rely on experts. To address these challenges, we present an automated, deep learning-based monitoring system that integrates YOLOv8, a state-of-the-art object detection algorithm, with DeepSORT, a robust multi-object tracking method, to identify and track coral formations in underwater video footage. Our system was fine-tuned using two curated and annotated datasets: AIMECORAL1 (580 images from the Southwest Indian Ocean) and AIMECORAL2 (282 images from New Caledonia, Pacific Ocean), encompassing diverse coral species and environmental conditions. The system's performance was evaluated using established metrics: object detection precision, Multiple Object Tracking Accuracy (MOTA), Multiple Object Tracking Precision (MOTP), and Identity F1 Score (IDF1). Precision improved from 59.9 % (after fine-tuning on AIMECORAL1) to 84.7 % on the combined datasets. The tracking system achieved a MOTA of 82.63 %, MOTP of 83.28 %, and IDF1 of 70.76 %, demonstrating reliable multi-object tracking in complex underwater environments. We applied our framework to a case study involving video transects from an outer reef site in New Caledonia, comparing data from 2021 and 2022. This automated solution offers a scalable, cost-effective alternative to traditional monitoring methods, supporting seamless, large-scale reef assessment. By leveraging deep learning, our approach enables more efficient data collection, contributing to the protection of these vulnerable ecosystems in the face of increasing environmental pressures.http://www.sciencedirect.com/science/article/pii/S1574954125001797Coral reefs ecosystemsDeep learningMethodologyYOLOv8DeepSORT |
| spellingShingle | Younes Ouassine Noël Conruyt Mohsen Kayal Philippe A. Martin Lionel Bigot Vignes Lebbe Regine Hajar Moussanif Jihad Zahir Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking Ecological Informatics Coral reefs ecosystems Deep learning Methodology YOLOv8 DeepSORT |
| title | Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking |
| title_full | Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking |
| title_fullStr | Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking |
| title_full_unstemmed | Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking |
| title_short | Deep learning for automated coral reef monitoring a novel system based on YOLOv8 detection and DeepSORT tracking |
| title_sort | deep learning for automated coral reef monitoring a novel system based on yolov8 detection and deepsort tracking |
| topic | Coral reefs ecosystems Deep learning Methodology YOLOv8 DeepSORT |
| url | http://www.sciencedirect.com/science/article/pii/S1574954125001797 |
| work_keys_str_mv | AT younesouassine deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT noelconruyt deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT mohsenkayal deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT philippeamartin deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT lionelbigot deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT vigneslebberegine deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT hajarmoussanif deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking AT jihadzahir deeplearningforautomatedcoralreefmonitoringanovelsystembasedonyolov8detectionanddeepsorttracking |