{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,28]],"date-time":"2026-02-28T04:29:55Z","timestamp":1772252995878,"version":"3.50.1"},"reference-count":30,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2021,8,29]],"date-time":"2021-08-29T00:00:00Z","timestamp":1630195200000},"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>In this paper, we examine two strategies for boosting the performance of ensembles of Siamese networks (SNNs) for image classification using two loss functions (Triplet and Binary Cross Entropy) and two methods for building the dissimilarity spaces (FULLY and DEEPER). With FULLY, the distance between a pattern and a prototype is calculated by comparing two images using the fully connected layer of the Siamese network. With DEEPER, each pattern is described using a deeper layer combined with dimensionality reduction. The basic design of the SNNs takes advantage of supervised k-means clustering for building the dissimilarity spaces that train a set of support vector machines, which are then combined by sum rule for a final decision. The robustness and versatility of this approach are demonstrated on several cross-domain image data sets, including a portrait data set, two bioimage and two animal vocalization data sets. Results show that the strategies employed in this work to increase the performance of dissimilarity image classification using SNN are closing the gap with standalone CNNs. Moreover, when our best system is combined with an ensemble of CNNs, the resulting performance is superior to an ensemble of CNNs, demonstrating that our new strategy is extracting additional information.<\/jats:p>","DOI":"10.3390\/s21175809","type":"journal-article","created":{"date-parts":[[2021,8,31]],"date-time":"2021-08-31T22:58:15Z","timestamp":1630450695000},"page":"5809","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Closing the Performance Gap between Siamese Networks for Dissimilarity Image Classification and Convolutional Neural Networks"],"prefix":"10.3390","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3502-7209","authenticated-orcid":false,"given":"Loris","family":"Nanni","sequence":"first","affiliation":[{"name":"Department of Information Engineering (DEI), University of Padova, 35131 Padova, Italy"}]},{"given":"Giovanni","family":"Minchio","sequence":"additional","affiliation":[{"name":"Department of Information Engineering (DEI), University of Padova, 35131 Padova, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7664-6930","authenticated-orcid":false,"given":"Sheryl","family":"Brahnam","sequence":"additional","affiliation":[{"name":"Department of Information Technology and Cybersecurity, Missouri State University, 901 S, National Street, Springfield, MO 65804, USA"}]},{"given":"Davide","family":"Sarraggiotto","sequence":"additional","affiliation":[{"name":"Department of Information Engineering (DEI), University of Padova, 35131 Padova, Italy"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0290-7354","authenticated-orcid":false,"given":"Alessandra","family":"Lumini","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering (DISI), University of Bologna, Via dell\u2019Universit\u00e0 50, 47521 Cesena, Italy"}]}],"member":"1968","published-online":{"date-parts":[[2021,8,29]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"P\u0119kalska, E., and Duin, R.P. (2005). The Dissimilarity Representation for Pattern Recognition\u2014Foundations and Applications, World Scientific.","DOI":"10.1142\/9789812703170"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Cha, S., and Srihari, S. (2000, January 1). Writer Identification: Statistical Analysis and Dichotomizer. Proceedings of the SSPR\/SPR, Alicante, Spain.","DOI":"10.1007\/3-540-44522-6_13"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Oliveira, L., Justino, E., and Sabourin, R. (2007, January 29). Off-line Signature Verification Using Writer-Independent Approach. Proceedings of the 2007 International Joint Conference on Neural Networks, Orlando, FL, USA.","DOI":"10.1109\/IJCNN.2007.4371358"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"213","DOI":"10.1007\/s10032-011-0166-4","article-title":"Writer verification using texture-based features","volume":"15","author":"Hanusiak","year":"2011","journal-title":"Int. J. Doc. Anal. Recognit."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"187","DOI":"10.1016\/j.ecoinf.2018.08.007","article-title":"Bird species identification using spectrogram and dissimilarity approach","volume":"48","author":"Zottesso","year":"2018","journal-title":"Ecol. Inform."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Souza, V.L.F., Oliveira, A., and Sabourin, R. (2018, January 22\u201325). A Writer-Independent Approach for Offline Signature Verification using Deep Convolutional Neural Networks Features. Proceedings of the 2018 7th Brazilian Conference on Intelligent Systems, S\u00e3o Paulo, Brazil.","DOI":"10.1109\/BRACIS.2018.00044"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"943","DOI":"10.1016\/S0167-8655(02)00024-7","article-title":"Dissimilarity representations allow for building good classifiers","volume":"23","author":"Duin","year":"2002","journal-title":"Pattern Recognit. Lett."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Nguyen, G., Worring, M., and Smeulders, A. (2006, January 26\u201327). Similarity learning via dissimilarity space in CBIR. Proceedings of the MIR\u201906, Santa Barbara, CA, USA.","DOI":"10.1145\/1178677.1178695"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"2367","DOI":"10.1016\/j.patcog.2013.09.026","article-title":"HEp-2 cells classification via sparse representation of textural features fused into dissimilarity space","volume":"47","author":"Theodorakopoulos","year":"2014","journal-title":"Pattern Recognit."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Hern\u00e1ndez-Dur\u00e1n, M., Cala\u00f1a, Y.P., and Vazquez, H.M. (2018, January 7\u201310). Low-Resolution Face Recognition with Deep Convolutional Features in the Dissimilarity Space. Proceedings of the IWAIPR, Chiang Mai, Thailand.","DOI":"10.1007\/978-3-030-01132-1_11"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Mekhazni, D., Bhuiyan, A., Ekladious, G., and Granger, \u00c9. (2020, January 23). Unsupervised Domain Adaptation in the Dissimilarity Space for Person Re-identification. Proceedings of the ECCV, Glasgow, Scotland.","DOI":"10.1007\/978-3-030-58583-9_10"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Nanni, L., Rigo, A., Lumini, A., and Brahnam, S. (2020). Spectrogram classification using dissimilarity space. Sensors, 10.","DOI":"10.3390\/app10124176"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Nanni, L., Brahnam, S., Lumini, A., and Maguolo, G. (2020). Animal sound classification using dissimilarity spaces. Appl. Sci., 10.","DOI":"10.20944\/preprints202010.0526.v1"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Nanni, L., Minchio, G., Brahnam, S., Maguolo, G., and Lumini, A. (2021). Experiments of image classification using dissimilarity spaces built with siamese networks. Sensors, 21.","DOI":"10.3390\/s21051573"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2783","DOI":"10.1007\/s10462-019-09746-z","article-title":"The dissimilarity approach: A review","volume":"53","author":"Costa","year":"2019","journal-title":"Artif. Intell. Rev."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Cartwright, H. (2020). Siamese neural networks: An overview. Artificial Neural Networks. Methods in Molecular Biology, Springer Protocols.","DOI":"10.1007\/978-1-0716-0826-5"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"669","DOI":"10.1142\/S0218001493000339","article-title":"Signature Verification Using A \u201cSiamese\u201d Time Delay Neural Network","volume":"7","author":"Bromley","year":"1993","journal-title":"Int. J. Pattern Recognit. Artif. Intell."},{"key":"ref_18","unstructured":"Agrawal, A. (2019). Dissimilarity learning via Siamese network predicts brain imaging data. arXiv, Available online: https:\/\/arxiv.org\/ftp\/arxiv\/papers\/1907\/1907.02591.pdf."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"San Biagio, M., Crocco, M., Cristani, M., Martelli, S., and Murino, V. (2013, January 1\u20138). Heterogeneous auto-similarities of characteristics (hasc): Exploiting relational information for classification. Proceedings of the IEEE Computer Vision (ICCV13), Sydney, Australia.","DOI":"10.1109\/ICCV.2013.105"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"2174","DOI":"10.1109\/78.157218","article-title":"Fast algorithms for the discrete cosine transform","volume":"49","author":"Feig","year":"1992","journal-title":"IEEE Trans. Signal. Process."},{"key":"ref_21","first-page":"6980","article-title":"Adam: A Method for Stochastic Optimization","volume":"1412","author":"Kingma","year":"2015","journal-title":"CoRR"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"4500","DOI":"10.1109\/TNNLS.2019.2955777","article-title":"diffGrad: An Optimization Method for Convolutional Neural Networks","volume":"31","author":"Dubey","year":"2020","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_23","unstructured":"Nanni, L., Maguolo, G., and Lumini, A. (2021, August 25). Exploiting Adam-like Optimization Algorithms to Improve the Performance of Convolutional Neural Networks. arXiv, Available online: https:\/\/arxiv.org\/ftp\/arxiv\/papers\/2103\/2103.14689.pdf."},{"key":"ref_24","unstructured":"You, K., Long, M., and Jordan, M.I. (2021, August 25). How Does Learning Rate Decay Help Modern Neural Networks. arXiv, Available online: https:\/\/arxiv.org\/abs\/1908.01878."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Zhang, S.-H., Zhao, Z., Xu, Z., Bellisario, K., and Pijanowski, B.C. (2018, January 15\u201320). Automatic Bird Vocalization Identification Based on Fusion of Spectral Pattern and Texture Features. Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal, Calgary, AB, Canada.","DOI":"10.1109\/ICASSP.2018.8462156"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Pandeya, Y.R., Kim, D., and Lee, J. (2018). Domestic cat sound classification using learned features from deep neural nets. Appl. Sci., 8.","DOI":"10.3390\/app8101949"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"154","DOI":"10.5391\/IJFIS.2018.18.2.154","article-title":"Domestic Cat Sound Classification Using Transfer Learning","volume":"18","author":"Pandeya","year":"2018","journal-title":"Int. J. Fuzzy Logic. Intell. Syst."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1016\/j.cmpb.2018.01.030","article-title":"Learning-based classification of informative laryngoscopic frames","volume":"158","author":"Moccia","year":"2018","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Nanni, L., Paci, M.P., Santos, F.L.C.d., Skottman, H., Juuti-Uusitalo, K., and Hyttinen, J. (2016). Texture descriptors ensembles enable image-based classification of maturation of human stem cell-derived retinal pigmented epithelium. PLoS ONE, 11.","DOI":"10.1371\/journal.pone.0149399"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"104121","DOI":"10.1016\/j.imavis.2021.104121","article-title":"Novel features for art movement classification of portrait paintings","volume":"108","author":"Liu","year":"2021","journal-title":"Image Vision Comput."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/17\/5809\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T06:54:49Z","timestamp":1760165689000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/21\/17\/5809"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,8,29]]},"references-count":30,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2021,9]]}},"alternative-id":["s21175809"],"URL":"https:\/\/doi.org\/10.3390\/s21175809","relation":{"has-preprint":[{"id-type":"doi","id":"10.20944\/preprints202108.0094.v1","asserted-by":"object"}]},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,8,29]]}}}