{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T01:54:59Z","timestamp":1774317299834,"version":"3.50.1"},"reference-count":39,"publisher":"MDPI AG","issue":"11","license":[{"start":{"date-parts":[[2022,5,28]],"date-time":"2022-05-28T00:00:00Z","timestamp":1653696000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"R\u00e9seau Qu\u00e9bec Maritime and Minist\u00e8re de l\u2019\u00c9conomie et de l\u2019Innovation du Qu\u00e9bec","award":["2017-2022-39557"],"award-info":[{"award-number":["2017-2022-39557"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>A key aspect of ocean protection consists in estimating the abundance of marine mammal population density within their habitat, which is usually accomplished using visual inspection and cameras from line-transect ships, small boats, and aircraft. However, marine mammal observation through vessel surveys requires significant workforce resources, including for the post-processing of pictures, and is further challenged due to animal bodies being partially hidden underwater, small-scale object size, occlusion among objects, and distracter objects (e.g., waves, sun glare, etc.). To relieve the human expert\u2019s workload while improving the observation accuracy, we propose a novel system for automating the detection of beluga whales (Delphinapterus leucas) in the wild from pictures. Our system relies on a dataset named Beluga-5k, containing more than 5.5 thousand pictures of belugas. First, to improve the dataset\u2019s annotation, we have designed a semi-manual strategy for annotating candidates in images with single (i.e., one beluga) and multiple (i.e., two or more belugas) candidate subjects efficiently. Second, we have studied the performance of three off-the-shelf object-detection algorithms, namely, Mask-RCNN, SSD, and YOLO v3-Tiny, on the Beluga-5k dataset. Afterward, we have set YOLO v3-Tiny as the detector, integrating single- and multiple-individual images into the model training. Our fine-tuned CNN-backbone detector trained with semi-manual annotations is able to detect belugas despite the presence of distracter objects with high accuracy (i.e., 97.05 mAP@0.5). Finally, our proposed method is able to detect overlapped\/occluded multiple individuals in images (beluga whales that swim in groups). For instance, it is able to detect 688 out of 706 belugas encountered in 200 multiple images, achieving 98.29% precision and 99.14% recall.<\/jats:p>","DOI":"10.3390\/s22114107","type":"journal-article","created":{"date-parts":[[2022,5,31]],"date-time":"2022-05-31T02:30:06Z","timestamp":1653964206000},"page":"4107","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Machine-Learning Approach for Automatic Detection of Wild Beluga Whales from Hand-Held Camera Pictures"],"prefix":"10.3390","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7103-6882","authenticated-orcid":false,"given":"Voncarlos M.","family":"Ara\u00fajo","sequence":"first","affiliation":[{"name":"D\u00e9partement des Sciences Naturelles, Universit\u00e9 du Qu\u00e9bec en Outaouais, Ripon, QC J0V 1V0, Canada"}]},{"given":"Ankita","family":"Shukla","sequence":"additional","affiliation":[{"name":"School of Arts, Media and Engineering, Arizona State University, Tempe, AZ 85281, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9618-0074","authenticated-orcid":false,"given":"Cl\u00e9ment","family":"Chion","sequence":"additional","affiliation":[{"name":"D\u00e9partement des Sciences Naturelles, Universit\u00e9 du Qu\u00e9bec en Outaouais, Ripon, QC J0V 1V0, Canada"}]},{"given":"S\u00e9bastien","family":"Gambs","sequence":"additional","affiliation":[{"name":"D\u00e9partement d\u2019Informatique, Universit\u00e9 du Qu\u00e9bec \u00e0 Montr\u00e9al (UQAM), Montreal, QC H2L 2C4, Canada"}]},{"given":"Robert","family":"Michaud","sequence":"additional","affiliation":[{"name":"Groupe de Recherche et d\u2019\u00c9ducation sur les Mammif\u00e8res Marins (GREMM), Tadoussac, QC G0T 2A0, Canada"}]}],"member":"1968","published-online":{"date-parts":[[2022,5,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/S0964-5691(00)00028-4","article-title":"The industrialisation of the world ocean","volume":"43","author":"Smith","year":"2000","journal-title":"Ocean. Coast. Manag."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1080\/13880290701229838","article-title":"Comparative Review of the Regional Marine Mammal Mitigation Guidelines Implemented During Industrial Seismic Surveys, and Guidance Towards a Worldwide Standard","volume":"10","author":"Weir","year":"2007","journal-title":"J. Int. Wildl. Law Policy"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1016\/j.marpolbul.2015.08.045","article-title":"JNCC guidelines for minimising the risk of injury and disturbance to marine mammals from seismic surveys: We can do better","volume":"100","author":"Wright","year":"2015","journal-title":"Mar. Pollut. Bull."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1111\/j.1523-1739.2006.00338.x","article-title":"Bycatch of marine mammals in U.S. and global fisheries","volume":"20","author":"Read","year":"2006","journal-title":"Conserv. Biol. J. Soc. Conserv. Biol."},{"key":"ref_5","first-page":"409","article-title":"Effectiveness of Monitoring Marine Mammals during Marine Seismic Surveys off Northeast Brazil","volume":"11","author":"Parente","year":"2011","journal-title":"J. Integr. Coast. Zone Manag."},{"key":"ref_6","unstructured":"Harwood, L., and Joynt, A. (2009). Factors Influencing the Effectiveness of Marine Mammal Observers on Seismic Vessels, with Examples from the Canadian Beaufort Sea, Fisheries and Oceans Canada."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"271","DOI":"10.3389\/fmars.2018.00271","article-title":"Long-Term Photo-Identification Reveals the Population Dynamics and Strong Site Fidelity of Adult Whale Sharks to the Coastal Waters of Donsol, Philippines","volume":"5","author":"McCoy","year":"2018","journal-title":"Front. Mar. Sci."},{"key":"ref_8","unstructured":"Michaud, R. (2014). St. Lawrence Estuary Beluga (Delphinapterus leucas) Population Parameters Based on Photo-Identification Surveys, 1989\u20132012, Canadian Science Advisory Secretariat. DFO Canadian Science Advisory Secretariat Research Document 2013\/130."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1071\/AM14023","article-title":"The pitfalls of wildlife camera trapping as a survey tool in Australia","volume":"37","author":"Meek","year":"2015","journal-title":"Aust. Mammal."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"823","DOI":"10.1080\/01431160600746456","article-title":"A survey of image classification methods and techniques for improving classification performance","volume":"28","author":"Lu","year":"2007","journal-title":"Int. J. Remote Sens."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"298","DOI":"10.1111\/mms.12141","article-title":"Recommendations for photo-identification methods used in capture-recapture models with cetaceans","volume":"31","author":"Urian","year":"2014","journal-title":"Mar. Mammal Sci."},{"key":"ref_12","unstructured":"Pfahringer, B., and Renz, J. (2015). Automating Marine Mammal Detection in Aerial Images Captured during Wildlife Surveys: A Deep Learning Approach. AI 2015: Advances in Artificial Intelligence, Springer International Publishing."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Borowicz, A., Le, H., Humphries, G., Nehls, G., H\u00f6schle, C., Kosarev, V., and Lynch, H.J. (2019). Aerial-trained deep learning networks for surveying cetaceans from satellite imagery. PLoS ONE, 14.","DOI":"10.1371\/journal.pone.0212532"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"14259","DOI":"10.1038\/s41598-019-50795-9","article-title":"Whale counting in satellite and aerial images with deep learning","volume":"9","author":"Guirado","year":"2019","journal-title":"Sci. Rep."},{"key":"ref_15","unstructured":"Mitrovic, T., Xue, B., and Li, X. (2018). Shark Detection from Aerial Imagery Using Region-Based CNN, a Study. AI 2018: Advances in Artificial Intelligence, Springer International Publishing."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1016\/j.ecoinf.2019.01.006","article-title":"A SIFT-based software system for the photo-identification of the Risso\u2019s dolphin","volume":"50","author":"Dimauro","year":"2019","journal-title":"Ecol. Inform."},{"key":"ref_17","unstructured":"Mejias, L., Duclos, G., Hodgson, A., and Maire, F. (2013, January 23\u201326). Automated marine mammal detection from aerial imagery. Proceedings of the OCEANS 2013 MTS\/IEEE\u2014San Diego: An Ocean in Common, San Diego, CA, USA."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Karnowski, J., Hutchins, E., and Johnson, C. (2015, January 5\u20139). Dolphin detection and tracking. Proceedings of the 2015 IEEE Winter Conference on Applications of Computer Vision Workshops, WACVW 2015, Waikoloa, HI, USA.","DOI":"10.1109\/WACVW.2015.10"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Park, M., Yang, W., Cao, Z., Kang, B., Connor, D., and Lea, M.A. (2019). Marine Vertebrate Predator Detection and Recognition in Underwater Videos by Region Convolutional Neural Network. Pacific Rim Knowledge Acquisition Workshop, Springer.","DOI":"10.1007\/978-3-030-30639-7_7"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"He, K., Gkioxari, G., Doll\u00e1r, P., and Girshick, R. (2018). Mask R-CNN. arXiv.","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Girshick, R. (2015, January 11\u201318). Fast R-CNN. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.","DOI":"10.1109\/ICCV.2015.169"},{"key":"ref_22","first-page":"91","article-title":"Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks","volume":"Volume 1","author":"Ren","year":"2015","journal-title":"Proceedings of the 28th International Conference on Neural Information Processing Systems"},{"key":"ref_23","unstructured":"Pedersen, M., Haurum, J.B., Gade, R., Moeslund, T.B., and Madsen, N. (2019, January 16\u201317). Detection of marine animals in a new underwater dataset with varying visibility. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Dawkins, M., Sherrill, L., Fieldhouse, K., Hoogs, A., Richards, B., Zhang, D., Prasad, L., Williams, K., Lauffenburger, N., and Wang, G. (2017, January 24\u201331). An Open-Source Platform for Underwater Image and Video Analytics. Proceedings of the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV), Santa Rosa, CA, USA.","DOI":"10.1109\/WACV.2017.105"},{"key":"ref_25","unstructured":"Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Fleet, D., Pajdla, T., Schiele, B., and Tuytelaars, T. (2014, January 6\u201312). Microsoft COCO: Common Objects in Context. Proceedings of the Computer Vision\u2014ECCV 2014, Zurich, Switzerland.","DOI":"10.1007\/978-3-319-10599-4"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1139\/juvs-2021-0024","article-title":"Detection and tracking of belugas, kayaks and motorized boats in drone video using deep learning","volume":"10","author":"Harasyn","year":"2022","journal-title":"Drone Syst. Appl."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016, January 8\u201316). SSD: Single Shot MultiBox Detector. Proceedings of the ECCV 2016, Amsterdam, The Netherlands.","DOI":"10.1007\/978-3-319-46448-0_2"},{"key":"ref_29","unstructured":"Skalski, P. (2021, October 05). Make Sense. Available online: https:\/\/github.com\/SkalskiP\/make-sense\/."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"3212","DOI":"10.1109\/TNNLS.2018.2876865","article-title":"Object Detection with Deep Learning: A Review","volume":"30","author":"Zhao","year":"2019","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. arXiv.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Ren, S., He, K., Girshick, R., and Sun, J. (2016). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. arXiv.","DOI":"10.1109\/TPAMI.2016.2577031"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep Residual Learning for Image Recognition. arXiv.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_34","unstructured":"Simonyan, K., and Zisserman, A. (2015). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. arXiv.","DOI":"10.1109\/CVPR.2016.91"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Xu, D., and Wu, Y. (2020). Improved YOLO-V3 with DenseNet for Multi-Scale Remote Sensing Target Detection. Sensors, 20.","DOI":"10.3390\/s20154276"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Hosang, J.H., Benenson, R., and Schiele, B. (2017). Learning non-maximum suppression. arXiv.","DOI":"10.1109\/CVPR.2017.685"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"4522","DOI":"10.1093\/bioinformatics\/btz259","article-title":"Biomedical image augmentation using Augmentor","volume":"35","author":"Bloice","year":"2019","journal-title":"Bioinformatics"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Lechevallier, Y., and Saporta, G. (2010, January 22\u201327). Large-Scale Machine Learning with Stochastic Gradient Descent. Proceedings of the COMPSTAT\u20192010, Paris, France.","DOI":"10.1007\/978-3-7908-2604-3"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/11\/4107\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T23:20:34Z","timestamp":1760138434000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/22\/11\/4107"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,28]]},"references-count":39,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2022,6]]}},"alternative-id":["s22114107"],"URL":"https:\/\/doi.org\/10.3390\/s22114107","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,5,28]]}}}