{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T17:07:48Z","timestamp":1790096868186,"version":"4.0.1"},"reference-count":46,"publisher":"Acoustical Society of America (ASA)","issue":"3","funder":[{"name":"Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences","award":["E4710401"],"award-info":[{"award-number":["E4710401"]}],"id":[{"id":"https:\/\/ror.org\/034t30j35","id-type":"ROR","asserted-by":"crossref"}]},{"name":"Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences","award":["E31X1203"],"award-info":[{"award-number":["E31X1203"]}],"id":[{"id":"https:\/\/ror.org\/034t30j35","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["pubs.aip.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:p>Vocalizations of coral reef fishes provide critical insights into population dynamics, community structure, and ecological processes, making passive acoustic monitoring an essential tool for coral reef conservation and management. However, automatic recognition of coral reef fish sounds remains highly challenging due to strong background noise, pronounced intra-class variability, and high inter-class similarity. To address these challenges, a contrastive reef-acoustic learning network (CoRAL-Net) is proposed for coral reef fish sound classification, in which a robust spectro-temporal feature extractor based on a convolutional neural network\u2013multi-head self-attention (CNN-MSA) module is designed to process the time\u2013frequency spectrograms. Furthermore, to improve triplet construction in contrastive learning, a confidence-aware sampling strategy is proposed to adaptively form triplets based on attention-derived uncertainty and embedding strength, thereby enhancing the representational capability of the CNN-MSA extractor through informative and class-balanced sample selection. This study validates CoRAL-Net on a predominantly field-collected dataset from coral reefs in Hainan Province, China, which contains three dominant types of fish sounds and natural noise recorded under challenging acoustic environments. Experimental results demonstrate that CoRAL-Net significantly outperforms the baseline methods in overall classification performance, highlighting the potential of this confidence-aware strategy to advance passive acoustic biodiversity monitoring in coral reef ecosystems.<\/jats:p>","DOI":"10.1121\/10.0046476","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T16:11:57Z","timestamp":1790093517000},"page":"2417-2427","update-policy":"https:\/\/doi.org\/10.1063\/aip-crossmark-policy-page","source":"Crossref","is-referenced-by-count":0,"title":["A confidence-aware contrastive deep learning network for coral reef fish sound classification"],"prefix":"10.1121","volume":"160","author":[{"given":"Tianze","family":"Hu","sequence":"first","affiliation":[{"name":"State Key Laboratory of Deep-Sea Science and Intelligent Technology, Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences 1 , Sanya 572000,","place":["China"]},{"name":"University of Chinese Academy of Sciences 2 , Beijing 100049,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ben","family":"Liu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Deep-Sea Science and Intelligent Technology, Institute of Deep-sea Science and Engineering, Chinese Academy of Sciences 1 , Sanya 572000,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaolei","family":"Yu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Tropical Oceanography, Guangdong Provincial Key Laboratory of Applied Marine Biology, South China Sea Institute of Oceanology, Chinese Academy of Sciences 3 , Guangzhou 510301,","place":["China"]},{"name":"Sanya Joint Laboratory of Marine Science Research, Key Laboratory of Tropical Marine Biotechnology of Hainan Province, Sanya Institute of Ocean Eco-Environmental Engineering, SCSIO 4 , Sanya 572000,","place":["China"]},{"name":"Sanya National Marine Ecosystem Research Station, Tropical Marine Biological Research Station in Hainan, Chinese Academy of Sciences 5 , Sanya 572000,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hui","family":"Huang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Tropical Oceanography, Guangdong Provincial Key Laboratory of Applied Marine Biology, South China Sea Institute of Oceanology, Chinese Academy of Sciences 3 , Guangzhou 510301,","place":["China"]},{"name":"Sanya Joint Laboratory of Marine Science Research, Key Laboratory of Tropical Marine Biotechnology of Hainan Province, Sanya Institute of Ocean Eco-Environmental Engineering, SCSIO 4 , Sanya 572000,","place":["China"]},{"name":"Sanya National Marine Ecosystem Research Station, Tropical Marine Biological Research Station in Hainan, Chinese Academy of Sciences 5 , Sanya 572000,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"231","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"key":"2026092212115453100_c1","doi-asserted-by":"publisher","first-page":"827","DOI":"10.1038\/nature02691","article-title":"Confronting the coral reef crisis","volume":"429","year":"2004","journal-title":"Nature"},{"key":"2026092212115453100_c2","first-page":"1597","article-title":"A simple framework for contrastive learning of visual representations","year":"2020"},{"key":"2026092212115453100_c3","article-title":"An image is worth 16\u2009\u00d7\u200916 words: Transformers for image recognition at scale","year":"2021"},{"key":"2026092212115453100_c4","volume-title":"Survey Manual for Tropical Marine Resources","year":"1997","edition":"2"},{"key":"2026092212115453100_c5","first-page":"571","article-title":"AST: Audio spectrogram transformer","year":"2021"},{"key":"2026092212115453100_c6","first-page":"1021","article-title":"Contrastive loss based frame-wise feature disentanglement for polyphonic sound event detection","year":"2024"},{"key":"2026092212115453100_c7","first-page":"244","article-title":"Fish migration monitoring from audio detection with CNNs","year":"2021"},{"key":"2026092212115453100_c8","first-page":"1735","article-title":"Dimensionality reduction by learning an invariant mapping","year":"2006"},{"key":"2026092212115453100_c9","first-page":"770","article-title":"Deep residual learning for image recognition","year":"2016"},{"key":"2026092212115453100_c10","doi-asserted-by":"publisher","first-page":"347","DOI":"10.3389\/fmars.2019.00347","article-title":"Caribbean Sea soundscapes: Monitoring humpback whales, biological sounds, geological events, and anthropogenic impacts of vessel noise","volume":"6","year":"2019","journal-title":"Front. Mar. Sci."},{"key":"2026092212115453100_c11","unstructured":"Hermans, A., Beyer, L., and Leibe, B. (2017). \u201cIn defense of the triplet loss for person re-identification,\u201d arXiv:1703.07737."},{"key":"2026092212115453100_c12","first-page":"131","article-title":"CNN architectures for large-scale audio classification","year":"2017"},{"key":"2026092212115453100_c13","volume-title":"Methods for Ecological Monitoring of Coral Reefs, Version 1: A Resource for Managers","year":"2004"},{"key":"2026092212115453100_c14","first-page":"1314","article-title":"Searching for MobileNetV3","year":"2019"},{"key":"2026092212115453100_c15","doi-asserted-by":"publisher","first-page":"929","DOI":"10.1126\/science.1085046","article-title":"Climate change, human impacts, and the resilience of coral reefs","volume":"301","year":"2003","journal-title":"Science"},{"key":"2026092212115453100_c16","doi-asserted-by":"publisher","first-page":"93","DOI":"10.3354\/meps11382","article-title":"Coral reef species assemblages are associated with ambient soundscapes","volume":"533","year":"2015","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"2026092212115453100_c17","doi-asserted-by":"publisher","first-page":"1720","DOI":"10.1121\/1.2836780","article-title":"An ecological acoustic recorder (EAR) for long-term monitoring of biological and anthropogenic sounds on coral reefs and other marine habitats","volume":"123","year":"2008","journal-title":"J. Acoust. Soc. Am."},{"key":"2026092212115453100_c18","article-title":"Decoupled weight decay regularization","year":"2019"},{"key":"2026092212115453100_c19","doi-asserted-by":"publisher","first-page":"1665","DOI":"10.1121\/10.0035829","article-title":"Rapid detection of fish calls within diverse coral reef soundscapes using a convolutional neural network","volume":"157","year":"2025","journal-title":"J. Acoust. Soc. Am."},{"key":"2026092212115453100_c20","doi-asserted-by":"publisher","first-page":"17459","DOI":"10.1038\/s41598-017-15838-z","article-title":"Patterns of biophonic periodicity on coral reefs in the Great Barrier Reef","volume":"7","year":"2017","journal-title":"Sci. Rep."},{"key":"2026092212115453100_c21","doi-asserted-by":"publisher","first-page":"295","DOI":"10.1080\/09524622.2017.1344930","article-title":"Soundscape diversity in the Great Barrier Reef: Lizard Island, a case study","volume":"27","year":"2018","journal-title":"Bioacoustics"},{"key":"2026092212115453100_c22","doi-asserted-by":"publisher","first-page":"257","DOI":"10.1111\/2041-210X.12330","article-title":"Measuring acoustic habitats","volume":"6","year":"2015","journal-title":"Methods Ecol. Evol."},{"key":"2026092212115453100_c23","doi-asserted-by":"publisher","first-page":"1439995","DOI":"10.3389\/frsen.2024.1439995","article-title":"Automatic detection of unidentified fish sounds: A comparison of traditional machine learning with deep learning","volume":"5","year":"2024","journal-title":"Front. Remote Sens."},{"key":"2026092212115453100_c24","doi-asserted-by":"publisher","first-page":"197","DOI":"10.3354\/meps13912","article-title":"Machine learning analysis reveals relationship between pomacentrid calls and environmental cues","volume":"681","year":"2022","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"2026092212115453100_c25","doi-asserted-by":"publisher","first-page":"106600","DOI":"10.1016\/j.marenvres.2024.106600","article-title":"Acoustic fish community in the Madeira Archipelago (North Atlantic Ocean): Characterization of sound diversity and daily patterns","volume":"199","year":"2024","journal-title":"Mar. Environ. Res."},{"key":"2026092212115453100_c26","doi-asserted-by":"publisher","first-page":"124220","DOI":"10.1016\/j.eswa.2024.124220","article-title":"Advancements in preprocessing, detection and classification techniques for ecoacoustic data: A comprehensive review for large-scale passive acoustic monitoring","volume":"252","year":"2024","journal-title":"Expert Syst. Appl."},{"key":"2026092212115453100_c27","unstructured":"Nasiri, A., and Hu, J. (2021). \u201cSoundCLR: Contrastive learning of representations for improved environmental sound classification,\u201d arXiv:2103.01929."},{"key":"2026092212115453100_c28","doi-asserted-by":"publisher","first-page":"e20275","DOI":"10.1016\/j.heliyon.2023.e20275","article-title":"Systematic review of machine learning methods applied to ecoacoustics and soundscape monitoring","volume":"9","year":"2023","journal-title":"Heliyon"},{"key":"2026092212115453100_c29","doi-asserted-by":"publisher","first-page":"1390687","DOI":"10.3389\/frsen.2024.1390687","article-title":"Machine learning for efficient segregation and labeling of potential biological sounds in long-term underwater recordings","volume":"5","year":"2024","journal-title":"Front. Remote Sens."},{"key":"2026092212115453100_c30","first-page":"1557","article-title":"Understanding the feature norm for out-of-distribution detection","year":"2023"},{"key":"2026092212115453100_c31","doi-asserted-by":"publisher","first-page":"1065","DOI":"10.1214\/aoms\/1177704472","article-title":"On estimation of a probability density function and mode","volume":"33","year":"1962","journal-title":"Ann. Math. Statist."},{"key":"2026092212115453100_c32","doi-asserted-by":"publisher","first-page":"868","DOI":"10.1016\/j.ecolind.2010.11.005","article-title":"A new methodology to infer the singing activity of an avian community: The acoustic complexity index (ACI)","volume":"11","year":"2011","journal-title":"Ecol. Indic."},{"key":"2026092212115453100_c33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.35995\/jea7010001","article-title":"\u2018To be, or not to be\u2019: Critical assessment of the use of \u03b1-acoustic diversity indices to evaluate the richness and abundance of coastal marine fish sounds","volume":"7","year":"2023","journal-title":"J. Ecoacoust."},{"key":"2026092212115453100_c34","doi-asserted-by":"publisher","first-page":"241938","DOI":"10.1098\/rsos.241938","article-title":"Highlighting the resilience potential of marine protected areas in the face of coral bleaching with passive acoustic monitoring","volume":"12","year":"2025","journal-title":"R. Soc. Open Sci."},{"key":"2026092212115453100_c35","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: A graphical aid to the interpretation and validation of cluster analysis","volume":"20","year":"1987","journal-title":"J. Comput. Appl. Math."},{"key":"2026092212115453100_c36","doi-asserted-by":"publisher","first-page":"241","DOI":"10.3354\/meps09839","article-title":"Sound production as an indicator of red hind density at a spawning aggregation","volume":"462","year":"2012","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"2026092212115453100_c37","first-page":"815","article-title":"FaceNet: A unified embedding for face recognition and clustering","year":"2015"},{"key":"2026092212115453100_c38","article-title":"Posterior attention models for sequence to sequence learning","year":"2019"},{"key":"2026092212115453100_c39","article-title":"Very deep convolutional networks for large-scale image recognition","year":"2015"},{"key":"2026092212115453100_c40","doi-asserted-by":"publisher","first-page":"1137","DOI":"10.1093\/ije\/dyq191","article-title":"The proof and measurement of association between two things","volume":"39","year":"2010","journal-title":"Int. J. Epidemiol."},{"key":"2026092212115453100_c41","doi-asserted-by":"publisher","first-page":"553","DOI":"10.1007\/s00338-012-1007-8","article-title":"Soundscapes from a Tropical Eastern Pacific reef and a Caribbean Sea reef","volume":"32","year":"2013","journal-title":"Coral Reefs"},{"key":"2026092212115453100_c42","doi-asserted-by":"publisher","first-page":"e4065","DOI":"10.1371\/journal.pone.0004065","article-title":"Rapid acoustic survey for biodiversity appraisal","volume":"3","year":"2008","journal-title":"PLoS One"},{"key":"2026092212115453100_c43","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","year":"2008","journal-title":"J. Mach. Learn. Res."},{"key":"2026092212115453100_c44","first-page":"5998","article-title":"Attention is all you need","year":"2017"},{"key":"2026092212115453100_c45","doi-asserted-by":"publisher","first-page":"109","DOI":"10.3354\/meps14128","article-title":"Temperature mediates chorusing behaviour associated with spawning in the sciaenid Argyrosomus regius","volume":"697","year":"2022","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"2026092212115453100_c46","first-page":"5022","article-title":"Multi-similarity loss with general pair weighting for deep metric learning","year":"2019"}],"container-title":["The Journal of the Acoustical Society of America"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/3\/2417\/21148516\/2417_1_10.0046476.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.aip.org\/asa\/jasa\/article-pdf\/160\/3\/2417\/21148516\/2417_1_10.0046476.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T16:11:59Z","timestamp":1790093519000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.aip.org\/jasa\/article\/160\/3\/2417\/3405067\/A-confidence-aware-contrastive-deep-learning"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,1]]},"references-count":46,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1121\/10.0046476","relation":{},"ISSN":["1520-8524"],"issn-type":[{"value":"1520-8524","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,9,1]]}}}