{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,5]],"date-time":"2026-04-05T05:32:42Z","timestamp":1775367162548,"version":"3.50.1"},"reference-count":46,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2021,9,5]],"date-time":"2021-09-05T00:00:00Z","timestamp":1630800000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Features play a crucial role in computer vision. Initially designed to detect salient elements by means of handcrafted algorithms, features now are often learned using different layers in convolutional neural networks (CNNs). This paper develops a generic computer vision system based on features extracted from trained CNNs. Multiple learned features are combined into a single structure to work on different image classification tasks. The proposed system was derived by testing several approaches for extracting features from the inner layers of CNNs and using them as inputs to support vector machines that are then combined by sum rule. Several dimensionality reduction techniques were tested for reducing the high dimensionality of the inner layers so that they can work with SVMs. The empirically derived generic vision system based on applying a discrete cosine transform (DCT) separately to each channel is shown to significantly boost the performance of standard CNNs across a large and diverse collection of image data sets. In addition, an ensemble of different topologies taking the same DCT approach and combined with global mean thresholding pooling obtained state-of-the-art results on a benchmark image virus data set.<\/jats:p>","DOI":"10.3390\/jimaging7090177","type":"journal-article","created":{"date-parts":[[2021,9,6]],"date-time":"2021-09-06T13:15:56Z","timestamp":1630934156000},"page":"177","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":20,"title":["Deep Features for Training Support Vector Machines"],"prefix":"10.3390","volume":"7","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":"Stefano","family":"Ghidoni","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":"Information Technology and Cybersecurity (ITC), Missouri State University, 901 S National, Springfield, MO 65804, USA"}]}],"member":"1968","published-online":{"date-parts":[[2021,9,5]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1023\/B:VISI.0000029664.99615.94","article-title":"Distinctive image features from scale-invariant keypoints","volume":"60","author":"Lowe","year":"2004","journal-title":"Int. J. Comput. Vis."},{"key":"ref_2","first-page":"404","article-title":"SURF: Speeded up robust features","volume":"1","author":"Bay","year":"2006","journal-title":"Eur. Conf. Comput. Vis."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.neunet.2014.09.003","article-title":"Deep learning in neural networks: An overview","volume":"61","author":"Schmidhuber","year":"2015","journal-title":"Neural Netw."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Bora, K., Chowdhury, M., Mahanta, L.B., Kundu, M., and Das, A. (2016, January 18\u201322). Pap smear image classification using convolutional neural network. Proceedings of the Tenth Indian Conference on Computer Vision, Graphics and Image Processing, Chengdu, China. No. 55.","DOI":"10.1145\/3009977.3010068"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"5017","DOI":"10.1109\/TIP.2015.2475625","article-title":"Pcanet: A simple deep learning baseline for image classification?","volume":"24","author":"Chan","year":"2015","journal-title":"IEEE Trans. Image Process."},{"key":"ref_6","unstructured":"Yosinski, J., Clune, J., Bengio, Y., and Lipson, H. (2014). How Transferable are Features in Deep Neural Networks?. arXiv."},{"key":"ref_7","unstructured":"Athiwaratkun, B., and Kang, K. (2015). Feature representation in convolutional neural networks. arXiv."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Yang, B., Yan, B., Lei, B., and Li, S.Z. (2015, January 7\u201313). Convolutional channel features. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.","DOI":"10.1109\/ICCV.2015.18"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1016\/j.patcog.2016.01.007","article-title":"String representations and distances in deep Convolutional Neural Networks for image classification","volume":"54","author":"Barat","year":"2016","journal-title":"Pattern Recognit. Bioinform."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s13755-018-0057-x","article-title":"Transfer learning based histopathologic image classification for breast cancer detection","volume":"6","author":"Deniz","year":"2018","journal-title":"Health Inf. Sci. Syst."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Razavian, A.S., Azizpour, H., Sullivan, J., and Carlsson, S. (2014). CNN features off-the-shelf: An astounding baseline for recognition. arXiv.","DOI":"10.1109\/CVPRW.2014.131"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Cristianini, N., and Shawe-Taylor, J. (2000). An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods, Cambridge University Press.","DOI":"10.1017\/CBO9780511801389"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Cimpoi, M., Maji, S., and Vedaldi, A. (2014). Deep convolutional filter banks for texture recognition and segmentation. arXiv.","DOI":"10.1109\/CVPR.2015.7299007"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Gong, Y., Wang, L., Guo, R., and Lazebnik, S. (2014). Multi-scale orderless pooling of deep convolutional activation features. arXiv.","DOI":"10.1007\/978-3-319-10584-0_26"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, S. (2014). Spatial pyramid pooling in deep convolutional networks for visual recognition. Computer Vision\u2014ECCV 2014, Springer. LNCS 8691.","DOI":"10.1007\/978-3-319-10578-9_23"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Forcen, J.I., Pagola, M., Barrenechea, E., and Bustince, H. (2020). Co-occurrence of deep convolutional features for image search. Image Vis. Comput., 97.","DOI":"10.1016\/j.imavis.2020.103909"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2013). Rich feature hierarchies for accurate object detection and semantic segmentation. arXiv.","DOI":"10.1109\/CVPR.2014.81"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Huang, H., and Xu, K. (2019). Combing Triple-Part Features of Convolutional Neural Networks for Scene Classification in Remote Sensing. Remote Sens., 11.","DOI":"10.3390\/rs11141687"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"158","DOI":"10.1016\/j.patcog.2017.05.025","article-title":"Handcrafted vs non-handcrafted features for computer vision classification","volume":"71","author":"Nanni","year":"2017","journal-title":"Pattern Recognit."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1016\/j.ins.2020.09.058","article-title":"Analysis of activation maps through global pooling measurements for texture classification","volume":"555","author":"Condori","year":"2021","journal-title":"Inf. Sci."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Allem, J.-P., Unger, J., and Cruz, T. (2018). Automated Identification of Hookahs (Waterpipes) on Instagram: An Application in Feature Extraction Using Convolutional Neural Network and Support Vector Machine Classification. J. Med. Internet Res., 20.","DOI":"10.2196\/preprints.10513"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1680","DOI":"10.1016\/j.procs.2020.04.180","article-title":"Deep Learning based Feature Extraction for Texture Classification","volume":"171","author":"Simon","year":"2020","journal-title":"Procedia Comput. Sci."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1049\/iet-cps.2019.0069","article-title":"CNN-SVM: A classification method for fruit fly image with the complex background","volume":"5","author":"Peng","year":"2020","journal-title":"IET Cyper-Phys. Syst. Theory Appl."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1155\/2021\/6663641","article-title":"Facial Expression Recognition Algorithm Based on Fusion of Transformed Multilevel Features and Improved Weighted Voting SVM","volume":"2021","author":"Meng","year":"2021","journal-title":"Math. Probl. Eng."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Sahoo, J., Ari, S., and Patra, S.K. (2019, January 16\u201318). Hand Gesture Recognition Using PCA Based Deep CNN Reduced Features and SVM Classifier. Proceedings of the 2019 IEEE International Symposium on Smart Electronic Systems (iSES) (Formerly iNiS), Rourkela, India.","DOI":"10.1109\/iSES47678.2019.00056"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"1527","DOI":"10.1162\/neco.2006.18.7.1527","article-title":"A fast learning algorithm for deep belief nets","volume":"18","author":"Hinton","year":"2006","journal-title":"Neural Comput."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L., Li, K., and Fei-Fei, L. (2009, January 20\u201325). ImageNet: A large-scale hierarchical image database. Proceedings of the CVPR, Miami, FL, USA.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Szegedy, C., Ioffe, S., Vanhoucke, V., and Alemi, A. (2016). Inception-v4, Inception-Resnet and the Impact of Residual Connections on Learning, Cornell University. Available online: https:\/\/arxiv.org\/pdf\/1602.07261.pdf.","DOI":"10.1609\/aaai.v31i1.11231"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016, January 27\u201330). Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_30","first-page":"3","article-title":"Densely Connected Convolutional Networks","volume":"1","author":"Huang","year":"2017","journal-title":"CVPR"},{"key":"ref_31","unstructured":"Duda, R.O., and Hart, P.E. (1973). Pattern Classification and Scene Analysis, Academic Press."},{"key":"ref_32","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_33","unstructured":"Lumini, A., Nanni, L., and Maguolo, G. (2019). Deep learning for Plankton and Coral Classification. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"1213","DOI":"10.1093\/bioinformatics\/17.12.1213","article-title":"A neural network classifier capable of recognizing the patterns of all major subcellular structures in fluorescence microscope images of HeLa cells","volume":"17","author":"Boland","year":"2001","journal-title":"BioInformatics"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"943","DOI":"10.1007\/s11517-008-0380-5","article-title":"IICBU 2008: A proposed benchmark suite for biological image analysis","volume":"46","author":"Shamir","year":"2008","journal-title":"Med. Biol. Eng. Comput."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1371\/journal.pone.0185110","article-title":"Grading of invasive breast carcinoma through Grassmannian VLAD encoding","volume":"12","author":"Dimitropoulos","year":"2017","journal-title":"PLoS ONE"},{"key":"ref_37","first-page":"34502","article-title":"Confident texture-based laryngeal tissue classification for early stage diagnosis support","volume":"4","author":"Moccia","year":"2017","journal-title":"J. Med. Imaging (Bellingham)"},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Hamilton, N., Pantelic, R., Hanson, K., and Teasdale, R.D. (2007). Fast automated cell phenotype classification. BMC Bioinform., 8.","DOI":"10.1186\/1471-2105-8-110"},{"key":"ref_39","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","year":"2006","journal-title":"J. Mach. Learn. Res."},{"key":"ref_40","unstructured":"Kylberg, G., Uppstr\u00f6m, M., and Sintorn, I.-M. (2013, January 20\u201323). Virus texture analysis using local binary patterns and radial density profiles. Proceedings of the 18th Iberoamerican Congress on Pattern Recognition (CIARP), Havana, Cuba."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Nanni, L., Luca, E.D., and Facin, M.L. (2020). Deep learning and hand-crafted features for virus image classification. J. Imaging, 6.","DOI":"10.3390\/jimaging6120143"},{"key":"ref_42","unstructured":"Geus, A.R., Backes, A.R., and Souza, J.R. (2020, January 27\u201329). Variability Evaluation of CNNs using Cross-validation on Viruses Images. Proceedings of the VISIGRAPP, Valletta, Malta."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1007\/s40305-018-0212-8","article-title":"Latent Local Feature Extraction for Low-Resolution Virus Image Classification","volume":"8","author":"Wen","year":"2020","journal-title":"J. Oper. Res. Soc. China"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Backes, A.R., and Junior, J.J.M.S. (2020, January 1\u20133). Virus Classification by Using a Fusion of Texture Analysis Methods. Proceedings of the 2020 International Conference on Systems, Signals and Image Processing (IWSSIP), Niteroi, Brazil.","DOI":"10.1109\/IWSSIP48289.2020.9145325"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"25","DOI":"10.1016\/j.patrec.2016.04.022","article-title":"Virus image classification using multi-scale completed local binary pattern features extracted from filtered images by multi-scale principal component analysis","volume":"79","author":"Wen","year":"2016","journal-title":"Pattern Recognit. Lett."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.biosystemseng.2015.01.005","article-title":"Computer vision for virus image classification","volume":"138","author":"Paci","year":"2015","journal-title":"Biosyst. Eng."}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/7\/9\/177\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T06:56:49Z","timestamp":1760165809000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/7\/9\/177"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,9,5]]},"references-count":46,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2021,9]]}},"alternative-id":["jimaging7090177"],"URL":"https:\/\/doi.org\/10.3390\/jimaging7090177","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,9,5]]}}}