Early detection of marine bioinvasion by sun corals using YOLOv8

Abstract

Sun coral (Tubastraea spp.) is an invasive species that poses a considerable threat to coastal ecosystems. Therefore, early detection is essential for effective monitoring and mitigation of its negative impacts on marine biodiversity. This study presents a novel computer vision approach for automated early detection of invasive Tubastraea species in underwater images. We used the YOLOv8 object detection model, which was trained and validated on a manually annotated dataset augmented with synthetic images. The data augmentation addressed the challenge of limited training data that is prevalent in underwater environments. The model achieved performance metrics (in terms of precision accuracy, recall, mAP50, and F1 score) of over 90% and detected both open and closed coral stage classes. Test phase results were compared with expert validation, demonstrating the model’s effectiveness in rapid detection (16 ms) and its limitations in areas highly covered by Tubastraea. This study demonstrates the potential of deep learning with data augmentation to facilitate the rapid assessment of large image datasets in monitoring sun coral bioinvasion. This approach has the potential to assist managers, taxonomists, and other professionals in the control of invasive alien species.

References

[1] Abayomi-Alli OO, Damaševičius R, Misra S, Maskeliūnas R. Cassava disease recognition from low-quality images using enhanced data augmentation model and deep learning Expert Syst, 2021, 38(7. e12746

[2] Alshahrani A, Ali H, Saif E, Alsayed M, Alshareef F (2024) Classification of coral reef species using computer vision and deep learning techniques. Eng Technol Appl Sci Res 14(5):16478–16485. https://doi.org/10.48084/etasr.8044

[3] Aota T, Ashizawa K, Mori H, Toda M, Chiba S. Detection of Anolis carolinensis using drone images and a deep neural network: an effective tool for controlling invasive species Biol Invasions, 2021, 23 5): 1321-1327.

[4] Bastos N, Calazans SH, Altvater L, Neves EG, Trujillo AL, Sharp WC, et al.. Western Atlantic invasion of sun corals: incongruence between morphology and genetic delimitation among morphotypes in the genus Tubastraea Bull Mar Sci, 2022, 98: 187-210.

[5] Bloice MD, Stocker C, Holzinger A (2017) Augmentor: an image augmentation library for machine learning. Preprint at arXiv:1708.04680

[6] Braga MDA, Paiva SV, de Gurjão LM, Teixeira CEP, Gurgel ALAR, Pereira PHC, et al.. Retirement risks: invasive coral on old oil platform on the Brazilian equatorial continental shelf Mar Pollut Bull, 2021, 165. 112156

[7] Brancaccio GR, Lagraf D, Pimentel LO (2023) Controlled removal of sun coral in drilling units hull. In: Offshore Technology Conference Brasil, Rio de Janeiro, p D012S051R001. https://doi.org/10.4043/32730-ms

[8] Buslaev A, Iglovikov VI, Khvedchenya E, Parinov A, Druzhinin M, Kalinin AA. Albumentations: fast and flexible image augmentations Information, 2020, 11(2): 125.

[9] Chong WS, Akmal KF, Shah MD (2023) The synergy of remote sensing in marine invasion science. In: Shah MD et al (eds) Marine biotechnology: applications in food, drugs and energy. Springer, Singapore, pp 299–313. https://doi.org/10.1007/978-981-99-0624-6_14

[10] Cook J, Coutts A (2017) The growing role of underwater robotics as a first line of defence for protecting Australia’s marine ecosystems. In: Australasian Coasts & Ports 2017: Working with Nature, p 268

[11] Creed JC, Fenner D, Sammarco P, Cairns S, Capel K, Junqueira AOR, et al.. The invasion of the azooxanthellate coral Tubastraea (Scleractinia: Dendrophylliidae) throughout the world: history, pathways and vectors Biol Invasions, 2017, 19(1): 283-305.

[12] Creed JC, Junqueira AOR, Fleury BG, Mantelatto MC, Oigman-Pszczol SS. The Sun-Coral Project: the first social-environmental initiative to manage the biological invasion of Tubastraea spp. in Brazil Manag Biol Invasions, 2017, 8(2): 181-195.

[13] de Oliveira SM, Davis M, de Macêdo Carneiro PB. Northward range expansion of the invasive coral (Tubastraea tagusensis) in the southwestern Atlantic Mar Biodiv, 2016, 48(3): 1651-1654.

[14] de Paula AF, Creed JC. Two species of the coral Tubastraea (Cnidaria, Scleractinia) in Brazil: a case of accidental introduction Bull Mar Sci, 2004, 74(1): 175-183

[15] de Paula AF, de Oliveira Pires D, Creed JC (2014) Reproductive strategies of two invasive sun corals (Tubastraea spp.) in the southwestern Atlantic. J Mar Biol Assoc UK 94(3):481–492. https://doi.org/10.1017/s0025315413001446

[16] Dutra BSVM, Carlos-Júnior LA, Creed JC. When species become invasive research becomes problem oriented: a synthesis of knowledge of the stony coral Tubastraea Biol Invasions, 2023, 25(7): 2069-2088.

[17] Elias N (2023) Deep learning methodology for early detection and outbreak prediction of invasive species growth. In: 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, pp 6324–6332. https://doi.org/10.1109/wacv56688.2023.00627

[18] Fenner D. New observations on the stony coral (Scleractinia, Milleporidae, and Stylasteridae) species of Belize (Central America) and Cozumel (Mexico) Bull Mar Sci, 1999, 64(1): 143-154

[19] Fenner D. Biogeography of three Caribbean corals (Scleractinia) and the invasion of Tubastraea coccinea into the Gulf of Mexico Bull Mar Sci, 2001, 69(3): 1175-1189

[20] Fenner D, Banks K. Orange Cup Coral Tubastraea coccinea invades Florida and the Flower Garden Banks, Northwestern Gulf of Mexico Coral Reefs, 2004, 23: 505-507.

[21] Furtado DP, Vieira EA, Nascimento WF, Inagaki KY, Bleuel J, Alves MAZ, et al.. #DeOlhoNosCorais: a polygonal annotated dataset to optimize coral monitoring PeerJ, 2023, 11. e16219

[22] Gao L, Li XF, Kong FZ, Yu RC, Guo Y, Ren YB. AlgaeNet: a deep-learning framework to detect floating green algae from optical and SAR imagery IEEE J Sel Top Appl Earth Observ Remote Sens, 2022, 15: 2782-2796.

[23] Gayá-Vilar A, Abad-Uribarren A, Rodríguez-Basalo A, Ríos P, Cristobo J, Prado E. Deep learning based characterization of cold-water coral habitat at central Cantabrian Natura 2000 sites using YOLOv8 J Mar Sci Eng, 2024, 12(9): 1617.

[24] Gómez-Ríos A, Tabik S, Luengo J, Shihavuddin ASM, Herrera F. Coral species identification with texture or structure images using a two-level classifier based on Convolutional Neural Networks Knowledge-Based Syst, 2019, 184. 104891

[25] Gómez-Ríos A, Tabik S, Luengo J, Shihavuddin ASM, Krawczyk B, Herrera F. Towards highly accurate coral texture images classification using deep convolutional neural networks and data augmentation Expert Syst Appl, 2019, 118: 315-328.

[26] González-Rivero M, Beijbom O, Rodriguez-Ramirez A, Bryant DEP, Ganase A, Gonzalez-Marrero Y, et al.. Monitoring of coral reefs using artificial intelligence: a feasible and cost-effective approach Remote Sens, 2020, 12(3): 489.

[27] Goodfellow I, Bengio Y, Courville A Deep learning, 2016 Cambridge MIT press

[28] Gorro K, Ilano A, Ranolo E, Pineda H, Sintos C, Gorro AJ (2023) Coral detection in fluorescence images and videos using YOLOV3 and YOLOV5. In: 2023 International Conference on Business Analytics for Technology and Security (ICBATS), Dubai, pp 1–6. https://doi.org/10.1109/icbats57792.2023.10111194

[29] Hoeksema BW, ten Hove HA. The invasive sun coral Tubastraea coccinea hosting a native Christmas tree worm at Curaçao Dutch Caribbean Mar Biodivers, 2017, 47(1): 59-65.

[30] IMO (2011) Guidelines for the control and management of ships' biofouling. https://wwwcdn.imo.org/localresources/en/KnowledgeCentre/IndexofIMOResolutions/AssemblyDocuments/A.1052(27).pdf

[31] IMO (2012) Guidance for minimizing the transfer of invasive aquatic species through hull fouling on recreational vessels. https://wwwcdn.imo.org/localresources/en/OurWork/Environment/Documents/MEPC.1-Circ.792.pdf

[32] Jiang YY, Qu MJ, Chen Y. Coral Detection, Ranging, and Assessment (CDRA) algorithm-based automatic estimation of coral reef coverage Mar Environ Res, 2023, 191. 106157

[33] Jocher G, Chaurasia A, Qiu J (2023) YOLO by Ultralytics. https://docs.ultralytics.com/

[34] Khorasani M, Abdou M, Hernández Fernández J (2022) Getting started with streamlit. In: Web application development with streamlit. Apress, Berkeley, CA, pp 1–30. https://doi.org/10.1007/978-1-4842-8111-6_1

[35] Lazzaro L, Viciani D, Dell’Olmo L, Foggi B. Predicting risk of invasion in a Mediterranean island using niche modelling and valuable biota Plant Biosyst, 2017, 151(2): 361-370.

[36] Li JYQ, Duce S, Joyce KE, Xiang W. SeeCucumbers: using deep learning and drone imagery to detect sea cucumbers on coral reef flats Drones, 2021, 5(2): 28.

[37] López C, Clemente S, Moreno S, Ocaña O, Herrera R, Moro L et al (2019) Invasive Tubastraea spp. and Oculina patagonica and other introduced scleractinians corals in the Santa Cruz de Tenerife (Canary Islands) harbor: ecology and potential risks. Reg Stud Mar Sci 29:100713. https://doi.org/10.1016/j.rsma.2019.100713

[38] Lopez-Marcano S, Brown CJ, Sievers M, Connolly RM. The slow rise of technology: computer vision techniques in fish population connectivity Aquat Conserv-Mar Freshw Ecosyst, 2020, 31(1): 210-217.

[39] Lopez-Marcano S, Jinks EL, Buelow CA, Brown CJ, Wang DD, Kusy B, et al.. Automatic detection of fish and tracking of movement for ecology Ecol Evol, 2021, 11(12): 8254-8263.

[40] Lumini A, Nanni L, Maguolo G. Deep learning for plankton and coral classification Appl Comput Inf, 2023, 19(3/4): 265-283.

[41] Machado AA, Masi BP, Aguiar AA, Ozorio MEC, Salles CN, Hostim-Silva M, et al.. Rocky reef incursions: challenges faced by reef fishes in a Brazilian Hope Spot region Mar Pollut Bull, 2023, 193. 115240

[42] Mahmood A, Bennamoun M, An S, Sohel F, Boussaid F, Hovey R et al (2017) Deep learning for coral classification. In: Handbook of neural computation. Academic Press, pp 383–401. https://doi.org/10.1016/B978-0-12-811318-9.00021-1

[43] Martinez B, Reaser JK, Dehgan A, Zamft B, Baisch D, McCormick C, et al.. Technology innovation: advancing capacities for the early detection of and rapid response to invasive species Biol Invasions, 2020, 22(1): 75-100.

[44] Miranda R, Tagliafico A, Kelaher B, Mariano-Neto E, Barros F. Impact of invasive corals Tubastrea spp. on native coral recruitment Mar Ecol Prog Ser, 2018, 605: 125-133.

[45] Mizrahi D, Pereira SF, Navarrete SA, Flores AAV. Allelopathic effects on the sun-coral invasion: facilitation, inhibition and patterns of local biodiversity Mar Biol, 2017, 164: 139.

[46] Mondal T, Raghunathan C, Chandra K. Report on status of invasive Tubastraea coccinea Lesson, 1829 in Andaman and Nicobar Islands, India Indian J Geo-Mar Sci, 2018, 47(11): 2241-2247

[47] Muksit A, Hasan F, Emon MFHB, Haque MR, Anwary AR, Shatabda S. YOLO-Fish: a robust fish detection model to detect fish in realistic underwater environment Ecol Inf, 2022, 72. 101847

[48] Oraño JFV, Napala JJO, Maaghop JFO, Elecito JC (2023) Automated coral lifeform classification using YOLOv5: a deep learning approach. In: Kabassi K et al (eds) Lecture notes in networks and systems, vol 783. Springer, Cham, pp 13–22. https://doi.org/10.1007/978-3-031-44097-7_2

[49] Pathak AR, Pandey M, Rautaray S. Application of deep learning for object detection Procedia Comput Sci, 2018, 132: 1706-1717.

[50] Pedersen M, Haurum JB, Gade R, Moeslund TB (2019) Detection of marine animals in a new underwater dataset with varying visibility. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp 18–26

[51] Piechaud N, Howell KL. Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision Ecol Inf, 2022, 71. 101786

[52] Raphael A, Dubinsky Z, Iluz D, Benichou JIC, Netanyahu NS. Deep neural network recognition of shallow water corals in the Gulf of Eilat (Aqaba) Sci Rep, 2020, 10: 12959.

[53] Raphael A, Dubinsky Z, Iluz D, Netanyahu NS. Neural network recognition of marine benthos and corals Diversity, 2020, 12(1): 29.

[54] Reaser JK, Burgiel SW, Kirkey J, Brantley KA, Veatch SD, Burgos-Rodríguez J. The early detection of and rapid response (EDRR) to invasive species: a conceptual framework and federal capacities assessment Biol Invasions, 2019, 22(1): 1-19.

[55] Roy AM, Bhaduri J, Kumar T, Raj K. WilDect-YOLO: an efficient and robust computer vision-based accurate object localization model for automated endangered wildlife detection Ecol Inf, 2023, 75. 101919

[56] Rusli NN, Mohtar IA (2023) Stony coral species recognition system using deep learning. In: 2023 4th International Conference on Artificial Intelligence and Data Sciences (AiDAS), Ioph, pp 325–330. https://doi.org/10.1109/aidas60501.2023.10284608

[57] Saleh A, Sheaves M, Azghadi MR. Computer vision and deep learning for fish classification in underwater habitats: a survey Fish Fish, 2022, 23(4): 977-999.

[58] Santoso SA, Jaya I, Priandana K (2024) Optimizing coral fish detection: faster R-CNN, SSD MobileNet, YOLOv5 comparison. IJCCS 18(2):1–5. https://doi.org/10.22146/ijccs.95011

[59] Savio LAC, Dias GM, Leite KL, Godoi SN, Figueiroa AC, Neto GFO, et al.. Sun coral management effectiveness in a wildlife refuge from south-eastern Brazil Aquat Conserv-Mar Freshw Ecosyst, 2021, 31(10): 2830-2841.

[60] Schettini R, Corchs S. Underwater image processing: state of the art of restoration and image enhancement methods EURASIP J Adv Signal Proc, 2010, 2010. 746052

[61] Schneider S, Taylor GW, Linquist S, Kremer SC. Past, present and future approaches using computer vision for animal re-identification from camera trap data Methods Ecol Evol, 2019, 10(4): 461-470.

[62] Shihavuddin ASM, Gracias N, Garcia R, Gleason A, Gintert B. Image-based coral reef classification and thematic mapping Remote Sens, 2013, 5(4): 1809-1841.

[63] Silva AG, Carlos-Júnior LA, Sato CYS, Lages BG, Neres-Lima V, de Oliveira FMS et al (2022) Living with an enemy: invasive sun-coral (Tubastraea spp.) competing against sponges Desmapsamma anchorata in southeastern Brazil. Mar Environ Res 174:105559. https://doi.org/10.1016/j.marenvres.2022.105559

[64] Silva AG, Lima RP, Gomes AN, Fleury BG, Creed JC. Expansion of the invasive corals Tubastraea coccinea and Tubastraea tagusensis into the Tamoios Ecological Station Marine Protected Area Brazil Aquat Invasions, 2011, 6(Supplement 1): S105-S110.

[65] Silva R, Vinagre C, Kitahara MV, Acorsi IV, Mizrahi D, Flores AAV. Sun coral invasion of shallow rocky reefs: effects on mobile invertebrate assemblages in Southeastern Brazil Biol Invasions, 2019, 21(4): 1339-1350.

[66] Sutherland WJ Ecological census techniques: a handbook, 2006 Cambridge University Press.

[67] Tait LW, Bulleid J, Rodgers LP, Seaward K, Olsen L, Woods C, et al.. Towards remote surveillance of marine pests: a comparison between remote operated vehicles and diver surveys Front Mar Sci, 2023, 10: 1102506.

[68] Terven J, Córdova-Esparza DM, Romero-González JA. A comprehensive review of YOLO architectures in computer vision: from YOLOv1 to YOLOv8 and YOLO-NAS Mach Learn Knowl Extr, 2023, 5(4): 1680-1716.

[69] Vaughan TW, Wells JW (1943) Revision of the suborders families, and genera of the scleractinia. In: GSA special papers. Geological Society of America, pp 1–394. https://doi.org/10.1130/SPE44-p1

[70] Wang W, Sun YF, Gao W, Xu WK, Zhang YX, Huang DX. Quantitative detection algorithm for deep-sea megabenthic organisms based on improved YOLOv5 Front Mar Sci, 2024, 11: 1301024.

[71] Xu SB, Zhang MH, Song W, Mei HB, He Q, Liotta A. A systematic review and analysis of deep learning-based underwater object detection Neurocomputing, 2023, 527: 204-232.

[72] Yang X, Shu L, Chen JN, Ferrag MA, Wu J, Nurellari E, Huang K. A survey on smart agriculture: development modes, technologies, and security and privacy challenges IEEE/CAA J Automatica Sin, 2021, 8(2): 273-302.

[73] Younes O, Jihad Z, Noël C, Mohsen K, Philippe AM, Eric C et al (2024) Automatic coral detection with YOLO: a deep learning approach for efficient and accurate coral reef monitoring. In: Communications in computer and information science, vol 1948. Springer, Cham, pp 170–177. https://doi.org/10.1007/978-3-031-50485-3_16

RIGHTS & PERMISSIONS

The Author(s)