{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,25]],"date-time":"2026-01-25T01:07:38Z","timestamp":1769303258411,"version":"3.49.0"},"reference-count":50,"publisher":"MDPI AG","issue":"19","license":[{"start":{"date-parts":[[2021,10,7]],"date-time":"2021-10-07T00:00:00Z","timestamp":1633564800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Deep learning has been successfully applied to many classification problems including underwater challenges. However, a long-standing issue with deep learning is the need for large and consistently labeled datasets. Although current approaches in semi-supervised learning can decrease the required amount of annotated data by a factor of 10 or even more, this line of research still uses distinct classes. For underwater classification, and uncurated real-world datasets in general, clean class boundaries can often not be given due to a limited information content in the images and transitional stages of the depicted objects. This leads to different experts having different opinions and thus producing fuzzy labels which could also be considered ambiguous or divergent. We propose a novel framework for handling semi-supervised classifications of such fuzzy labels. It is based on the idea of overclustering to detect substructures in these fuzzy labels. We propose a novel loss to improve the overclustering capability of our framework and show the benefit of overclustering for fuzzy labels. We show that our framework is superior to previous state-of-the-art semi-supervised methods when applied to real-world plankton data with fuzzy labels. Moreover, we acquire 5 to 10% more consistent predictions of substructures.<\/jats:p>","DOI":"10.3390\/s21196661","type":"journal-article","created":{"date-parts":[[2021,10,10]],"date-time":"2021-10-10T21:37:49Z","timestamp":1633901869000},"page":"6661","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":11,"title":["Fuzzy Overclustering: Semi-Supervised Classification of Fuzzy Labels with Overclustering and Inverse Cross-Entropy"],"prefix":"10.3390","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6945-5957","authenticated-orcid":false,"given":"Lars","family":"Schmarje","sequence":"first","affiliation":[{"name":"Multimedia Information Processing Group, Kiel University, 24118 Kiel, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5118-145X","authenticated-orcid":false,"given":"Johannes","family":"Br\u00fcnger","sequence":"additional","affiliation":[{"name":"Multimedia Information Processing Group, Kiel University, 24118 Kiel, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4159-1367","authenticated-orcid":false,"given":"Monty","family":"Santarossa","sequence":"additional","affiliation":[{"name":"Multimedia Information Processing Group, Kiel University, 24118 Kiel, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6603-9907","authenticated-orcid":false,"given":"Simon-Martin","family":"Schr\u00f6der","sequence":"additional","affiliation":[{"name":"Multimedia Information Processing Group, Kiel University, 24118 Kiel, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7851-9107","authenticated-orcid":false,"given":"Rainer","family":"Kiko","sequence":"additional","affiliation":[{"name":"Laboratoire d\u2019Oc\u00e9anographie de Villefranche, Sorbonne Universit\u00e9, 06230 Villefranche-sur-Mer, France"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4398-1569","authenticated-orcid":false,"given":"Reinhard","family":"Koch","sequence":"additional","affiliation":[{"name":"Multimedia Information Processing Group, Kiel University, 24118 Kiel, Germany"}]}],"member":"1968","published-online":{"date-parts":[[2021,10,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"14671","DOI":"10.1038\/s41598-020-71639-x","article-title":"A realistic fish-habitat dataset to evaluate algorithms for underwater visual analysis","volume":"10","author":"Saleh","year":"2020","journal-title":"Sci. Rep."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"315","DOI":"10.1016\/j.eswa.2018.10.010","article-title":"Towards highly accurate coral texture images classification using deep convolutional neural networks and data augmentation","volume":"118","author":"Tabik","year":"2019","journal-title":"Expert Syst. Appl."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Thum, G.W., Tang, S.H., Ahmad, S.A., and Alrifaey, M. (2020). Toward a highly accurate classification of underwater cable images via deep convolutional neural network. J. Mar. Sci. Eng., 8.","DOI":"10.3390\/jmse8110924"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Knausg\u00e5rd, K.M., Wiklund, A., S\u00f8rdalen, T.K., Halvorsen, K.T., Kleiven, A.R., Jiao, L., and Goodwin, M. (2021). Temperate fish detection and classification: A deep learning based approach. Appl. Intell.","DOI":"10.1007\/s10489-020-02154-9"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"196","DOI":"10.3389\/fmars.2019.00196","article-title":"Globally consistent quantitative observations of planktonic ecosystems","volume":"6","author":"Lombard","year":"2019","journal-title":"Front. Mar. Sci."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"834","DOI":"10.3389\/fmars.2019.00834","article-title":"Sinking Organic Particles in the Ocean\u2014Flux Estimates From in situ Optical Devices","volume":"6","author":"Giering","year":"2020","journal-title":"Front. Mar. Sci."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"941","DOI":"10.1093\/icesjms\/fsx216","article-title":"A new wave of marine evidence-based management: Emerging challenges and solutions to transform monitoring, evaluating, and reporting","volume":"75","author":"Addison","year":"2018","journal-title":"ICES J. Mar. Sci."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"61","DOI":"10.3354\/meps11775","article-title":"Comparison of image annotation data generated by multiple investigators for benthic ecology","volume":"552","author":"Durden","year":"2016","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Schoening, T., Bergmann, M., Ontrup, J., Taylor, J., Dannheim, J., Gutt, J., Purser, A., and Nattkemper, T.W. (2012). Semi-automated image analysis for the assessment of megafaunal densities at the Artic deep-sea observatory HAUSGARTEN. PLoS ONE, 7.","DOI":"10.1371\/journal.pone.0038179"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Schr\u00f6der, S.M., Kiko, R., and Koch, R. (2020). MorphoCluster: Efficient Annotation of Plankton images by Clustering. Sensors, 20.","DOI":"10.3390\/s20113060"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"101759","DOI":"10.1016\/j.media.2020.101759","article-title":"Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis","volume":"65","author":"Karimi","year":"2020","journal-title":"Med. Image Anal."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1030","DOI":"10.1017\/S1751731118003038","article-title":"\u2018Tailception\u2019: Using neural networks for assessing tail lesions on pictures of pig carcasses","volume":"13","author":"Dippel","year":"2019","journal-title":"Animal"},{"key":"ref_13","first-page":"374","article-title":"2D and 3D Segmentation of Uncertain Local Collagen Fiber Orientations in SHG Microscopy","volume":"Volume 11824 LNCS","author":"Schmarje","year":"2019","journal-title":"DAGM German Conference of Pattern Regocnition"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1342","DOI":"10.1038\/s41591-018-0107-6","article-title":"Clinically applicable deep learning for diagnosis and referral in retinal disease","volume":"24","author":"Ledsam","year":"2018","journal-title":"Nat. Med."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"1413","DOI":"10.1109\/JBHI.2019.2944643","article-title":"Deep Learning-Based Gleason Grading of Prostate Cancer From Histopathology Images\u2014Role of Multiscale Decision Aggregation and Data Augmentation","volume":"24","author":"Karimi","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1016\/j.ebiom.2015.05.009","article-title":"Crowdsourcing the general public for large scale molecular pathology studies in cancer","volume":"2","author":"Lynn","year":"2015","journal-title":"EBioMedicine"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"17","DOI":"10.3354\/meps247017","article-title":"Do experts make mistakes? A comparison of human and machine identification of dinoflagellates","volume":"247","author":"Culverhouse","year":"2003","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Tarling, P., Cantor, M., Clap\u00e9s, A., and Escalera, S. (2021). Deep learning with self-supervision and uncertainty regularization to count fish in underwater images. arXiv.","DOI":"10.1371\/journal.pone.0267759"},{"key":"ref_19","unstructured":"Berthelot, D., Carlini, N., Cubuk, E.D., Kurakin, A., Sohn, K., Zhang, H., and Raffel, C. (2019). ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring. arXiv."},{"key":"ref_20","unstructured":"Zhai, X., Oliver, A., Kolesnikov, A., and Beyer, L. (November, January 27). S4L: Self-Supervised Semi-Supervised Learning. Proceedings of the IEEE International Conference on Computer Vision, Seoul, Korea."},{"key":"ref_21","unstructured":"Chen, T., Kornblith, S., Norouzi, M., and Hinton, G. (2020). A Simple Framework for Contrastive Learning of Visual Representations. arXiv."},{"key":"ref_22","unstructured":"Gidaris, S., Singh, P., and Komodakis, N. (2018). Unsupervised Representation Learning by Predicting Image Rotations. arXiv."},{"key":"ref_23","unstructured":"Ji, X., Henriques, J.F., Vedaldi, A., Ji, X., Henriques, J.F., and Vedaldi, A. (November, January 27). Invariant information clustering for unsupervised image classification and segmentation. Proceedings of the IEEE International Conference on Computer Vision, Seoul, Korea."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"82146","DOI":"10.1109\/ACCESS.2021.3084358","article-title":"A Survey on Semi-, Self-and Unsupervised Learning for Image Classification","volume":"9","author":"Schmarje","year":"2021","journal-title":"IEEE Access"},{"key":"ref_25","unstructured":"Coates, A., Ng, A., and Lee, H. (2011, January 11\u201313). An analysis of single-layer networks in unsupervised feature learning. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, Fort Lauderdale, FL, USA."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"106771","DOI":"10.1016\/j.knosys.2021.106771","article-title":"Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey","volume":"215","author":"Algan","year":"2021","journal-title":"Knowl.-Based Syst."},{"key":"ref_27","unstructured":"Song, H., Kim, M., Park, D., and Lee, J. (2020). Learning from Noisy Labels with Deep Neural Networks: A Survey. arXiv."},{"key":"ref_28","unstructured":"Nguyen, D.T., Mummadi, C.K., Ngo, T.P.N., Nguyen, T.H.P., Beggel, L., and Brox, T. (2019). SELF: Learning to Filter Noisy Labels with Self-Ensembling. arXiv."},{"key":"ref_29","unstructured":"Laine, S., and Aila, T. (2016). Temporal ensembling for semi-supervised learning. arXiv."},{"key":"ref_30","unstructured":"Li, J., Socher, R., and Hoi, S.C.H. (2020). DivideMix: Learning with Noisy Labels as Semi-supervised Learning. arXiv."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1734","DOI":"10.1109\/TKDE.2016.2545658","article-title":"Label distribution learning","volume":"28","author":"Geng","year":"2016","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"2825","DOI":"10.1109\/TIP.2017.2689998","article-title":"Deep Label Distribution Learning With Label Ambiguity","volume":"26","author":"Gao","year":"2017","journal-title":"IEEE Trans. Image Process."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1445","DOI":"10.1007\/s11063-019-10142-5","article-title":"Semi-supervised Fuzzy Min\u2013Max Neural Network for Data Classification","volume":"51","author":"Liu","year":"2020","journal-title":"Neural Process. Lett."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Kowsari, K., Bari, N., Vichr, R., and Goodarzi, F.A. (2018). FSL-BM: Fuzzy Supervised Learning with Binary Meta-Feature for Classification. Future of Information and Communication Conference, Springer.","DOI":"10.1007\/978-3-030-03405-4_46"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"El-Zahhar, M.M., and El-Gayar, N.F. (December, January 29). A semi-supervised learning approach for soft labeled data. Proceedings of the 2010 10th International Conference on Intelligent Systems Design and Applications, Cairo, Egypt.","DOI":"10.1109\/ISDA.2010.5687034"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Liu, Y., Liang, X., Tong, S., and Kumada, T. (2019, January 22\u201325). Photo Shot-Type Disambiguation by Multi-Classifier Semi-Supervised Learning. Proceedings of the 2019 IEEE International Conference on Image Processing (ICIP), Taipei, Taiwan.","DOI":"10.1109\/ICIP.2019.8803310"},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Caron, M., Bojanowski, P., Joulin, A., and Douze, M. (2018, January 8\u201314). Deep clustering for unsupervised learning of visual features. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.","DOI":"10.1007\/978-3-030-01264-9_9"},{"key":"ref_38","unstructured":"Sohn, K., Berthelot, D., Li, C.L., Zhang, Z., Carlini, N., Cubuk, E.D., Kurakin, A., Zhang, H., and Raffel, C. (2020). FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence. arXiv."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 8\u201316). Identity Mappings in Deep Residual Networks. Proceedings of the European Conference on Computer Vision (ECCV), Amsterdam, the Netherlands.","DOI":"10.1007\/978-3-319-46493-0_38"},{"key":"ref_40","first-page":"529","article-title":"Semi-supervised learning by entropy minimization","volume":"367","author":"Grandvalet","year":"2005","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_41","unstructured":"Xie, Q., Dai, Z., Hovy, E., Luong, M.T., and Le, Q.V. (2019). Unsupervised Data Augmentation for Consistency Training. arXiv."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"1979","DOI":"10.1109\/TPAMI.2018.2858821","article-title":"Virtual adversarial training: A regularization method for supervised and semi-supervised learning","volume":"41","author":"Miyato","year":"2018","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_43","unstructured":"Berthelot, D., Carlini, N., Goodfellow, I., Papernot, N., Oliver, A., and Raffel, C.A. (2019). Mixmatch: A holistic approach to semi-supervised learning. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"462","DOI":"10.4319\/lom.2010.8.462","article-title":"The Underwater Vision Profiler 5: An advanced instrument for high spatial resolution studies of particle size spectra and zooplankton","volume":"8","author":"Picheral","year":"2010","journal-title":"Limnol. Oceanogr. Methods"},{"key":"ref_45","unstructured":"Picheral, M., Colin, S., and Irisson, J.O. (2021, October 06). EcoTaxa, a Tool for the Taxonomic Classification of Images. Available online: https:\/\/ecotaxa.obs-vlfr.fr\/."},{"key":"ref_46","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2012","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_47","unstructured":"Krizhevsky, A., and Hinton, G. (2021, October 06). Learning Multiple Layers of Features from Tiny Images. Technical Report. Available online: https:\/\/www.cs.toronto.edu\/~kriz\/learning-features-2009-TR.pdf."},{"key":"ref_48","doi-asserted-by":"crossref","unstructured":"Van Gansbeke, W., Vandenhende, S., Georgoulis, S., Proesmans, M., and Van Gool, L. (2020). Scan: Learning to classify images without labels. European Conference on Computer Vision, Springer.","DOI":"10.1007\/978-3-030-58607-2_16"},{"key":"ref_49","unstructured":"Tarvainen, A., and Valpola, H. (2017). Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. arXiv."},{"key":"ref_50","first-page":"2","article-title":"Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks","volume":"Volume 3","author":"Lee","year":"2013","journal-title":"Workshop on Challenges in Representation Learning"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/19\/6661\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T07:10:09Z","timestamp":1760166609000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/19\/6661"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,10,7]]},"references-count":50,"journal-issue":{"issue":"19","published-online":{"date-parts":[[2021,10]]}},"alternative-id":["s21196661"],"URL":"https:\/\/doi.org\/10.3390\/s21196661","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,10,7]]}}}