Please use this identifier to cite or link to this item: https://rima.ufrrj.br/jspui/handle/20.500.14407/15771
Full metadata record
DC FieldValueLanguage
dc.contributor.authorTorres, Helainy Ignacio de Almeida-
dc.date.accessioned2024-01-23T12:05:36Z-
dc.date.available2024-01-23T12:05:36Z-
dc.date.issued2022-05-30-
dc.identifier.urihttps://rima.ufrrj.br/jspui/handle/20.500.14407/15771-
dc.description.abstractMachine learning (Aprendizado de Máquina) ganhou destaque por ser uma área da inteligên- cia artificial que permite que algoritmos tomem decisões através do conhecimento extraído de amostras de dados. Uma das áreas de Machine Learning são os algoritmos de classificação, que se baseiam em prever a classe de uma observação dada. Existem vários métodos na literatura, que resolvem problemas de classificação como Rede Neural, SVM, KNN entre outros. Uma das semelhanças entre eles é utilizar a métrica euclidiana para determinar erros e aproxima- ções. Nesse trabalho propomos construir um algoritmo baseado no KNN utilizando a métrica riemanniana para o problema de classificação de imagens. Os bancos de imagens utilizados durante a pesquisa são de imagens médicas e cada imagem será representada como uma matriz de covariância. O método proposto foi comparado com o KNN clássico que utiliza a métrica euclidiana e em todosos testes realizados se mostrou superior, apesar da qualidade das imagem, demonstrando que a técnica tem muito a oferecer.pt_BR
dc.description.sponsorshipCoordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPESpt_BR
dc.languageporpt_BR
dc.publisherUniversidade Federal Rural do Rio de Janeiropt_BR
dc.subjectAprendizado de Máquinapt_BR
dc.subjectMétrica Riemannianapt_BR
dc.subjectKNNpt_BR
dc.subjectMachine Learningpt_BR
dc.subjectRiemannian Metricspt_BR
dc.titleUma proposta do algoritmo KNN sobre uma perspectiva riemanniana para o problema de classificação de imagenspt_BR
dc.title.alternativeA proposal of th KNN algorithm on a riemannian perspective for the image classification problemen
dc.typeDissertaçãopt_BR
dc.description.abstractOtherMachine learning has gained prominence as an area of artificial intelligence that allows algorithms to make decisions through knowledge extracted from data samples. One of the areas of Machine learning is classification algorithms, which are based on predicting the class of a given observation. There are several methods in the literature that solve classification problems such as Neural Network, SVM, KNN, among others. One of the similarities between them is to use the Euclidean metric to determine errors and approximations. In this work we propose to build an algorithm based on KNN using the Riemannian metric for the image classification problem. The image banks used during the research are of medical images and each image will be represented as a covariance matrix. The proposed method was compared with the classical KNN that uses the Euclidean metric and in all tests performed it proved to be superior, despite the image quality, demonstrating that the technique has a lot to offer.pt_BR
dc.contributor.advisor1Cruz, Marcelo Dib-
dc.contributor.advisor1ID016.628.007-03pt_BR
dc.contributor.advisor1IDhttps://orcid.org/0000-0002-0380-144Xpt_BR
dc.contributor.advisor1Latteshttp://lattes.cnpq.br/7385995443437070pt_BR
dc.contributor.advisor-co1Gregório, Ronaldo Malheiros-
dc.contributor.advisor-co1ID077.117.167-61pt_BR
dc.contributor.advisor-co1IDhttps://orcid.org/0000-0003-2229-0523pt_BR
dc.contributor.advisor-co1Latteshttp://lattes.cnpq.br/4502104424266743pt_BR
dc.contributor.referee1Cruz, Marcelo Dib-
dc.contributor.referee1ID016.628.007-03pt_BR
dc.contributor.referee1IDhttps://orcid.org/0000-0002-0380-144Xpt_BR
dc.contributor.referee1Latteshttp://lattes.cnpq.br/7385995443437070pt_BR
dc.contributor.referee2Vera-Tudela, Carlos Andrés Reyna-
dc.contributor.referee2IDhttps://orcid.org/0000-0001-5855-8611pt_BR
dc.contributor.referee2Latteshttp://lattes.cnpq.br/6509989261742578pt_BR
dc.contributor.referee3França, Juliana Baptista dos Santos-
dc.contributor.referee3ID053.276.397-11pt_BR
dc.contributor.referee3Latteshttp://lattes.cnpq.br/9341068095520817pt_BR
dc.creator.ID092.252.057-75pt_BR
dc.creator.Latteshttp://lattes.cnpq.br/0381719929316735pt_BR
dc.publisher.countryBrasilpt_BR
dc.publisher.departmentInstituto de Ciências Exataspt_BR
dc.publisher.initialsUFRRJpt_BR
dc.publisher.programPrograma de Pós-Graduação em Modelagem Matemática e Computacionalpt_BR
dc.relation.references1 THIAGO, E. R. de S. Reconhecimento de voz utilizando extração de coeficientes mel-cepstrais e redes neurais artificiais. 2017. 2 TUZEL, O.; PORIKLI, F.; MEER, P. Pedestrian detection via classification on riemannian manifolds. IEEE transactions on pattern analysis and machine intelligence, IEEE, v. 30, n. 10, p. 1713–1727, 2008. 3 LAN, X.; MA, A. J.; YUEN, P. C. Multi-cue visual tracking using robust feature-level fusion based on joint sparse representation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. [S.l.: s.n.], 2014. p. 1194–1201. 4 LAN, X.; ZHANG, S.; YUEN, P. C. Robust joint discriminative feature learning for visual tracking. In: IJCAI. [S.l.: s.n.], 2016. p. 3403–3410. 5 CHAN, H.-P. et al. Improvement in radiologists’ detection of clustered microcalcifications on mammograms. Arbor, v. 1001, p. 48109–0326, 1990. 6 DOI, K. et al. Computer-aided diagnosis: present and future. a new horizon on medical physics and biomedical engineering. Elsevier Science Publishers BV, p. 59–66, 1991. 7 ELLIS, I. et al. Early experience in breast cancer screening: emphasis on development of protocols for triple assessment. The Breast, Elsevier, v. 2, n. 3, p. 148–153, 1993. 8 PETRICK, N. et al. An adaptive density-weighted contrast enhancement filter for mammographic breast mass detection. IEEE Transactions on Medical Imaging, IEEE, v. 15, n. 1, p. 59–67, 1996. 9 GIGER, M.; MACMAHON, H. Image processing and computer-aided diagnosis. Radiologic Clinics of North America, v. 34, n. 3, p. 565–596, 1996. 10 LAN, X. et al. Joint sparse representation and robust feature-level fusion for multi-cue visual tracking. IEEE Transactions on Image Processing, IEEE, v. 24, n. 12, p. 5826–5841, 2015. 11 LAN, X. et al. Learning common and feature-specific patterns: a novel multiple-sparse- representation-based tracker. IEEE Transactions on Image Processing, IEEE, v. 27, n. 4, p. 2022–2037, 2017. 12 DALAL, N.; TRIGGS, B. Histograms of oriented gradients for human detection. In: IEEE. 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR’05). [S.l.], 2005. v. 1, p. 886–893. 13 LOWE, D. G. Distinctive image features from scale-invariant keypoints. International journal of computer vision, Springer, v. 60, n. 2, p. 91–110, 2004. 14 TUZEL, O.; PORIKLI, F.; MEER, P. Region covariance: A fast descriptor for detection and classification. In: SPRINGER. European conference on computer vision. [S.l.], 2006. p. 589–600. 15 WANG, R. et al. Covariance discriminative learning: A natural and efficient approach to image set classification. In: IEEE. 2012 IEEE conference on computer vision and pattern recognition. [S.l.], 2012. p. 2496–2503. 16 GONZALEZ, R. C.; WOODS, R. C. Processamento digital de imagens . [S.l.]: Pearson Educación, 2009. 17 PEREIRA, A. S.; RAFAEL, J. A. Processamento de imagens em medicina. Acta Medica Portuguesa, 1992. 18 SILVA, A. M. M. da; PATROCÍNIO, A. C.; SCHIABEL, H. Processamento e análise de imagens médicas. REVISTA BRASILEIRA DE FÍSICA MÉDICA (ONLINE), 2019. 19 LEÃO, P. P. de S. et al. Detecção de covid-19 em imagens de raio-x utilizando redes convolucionais. Journal of Health Informatics, v. 12, 2021. 20 RICHARDS, J. A.; RICHARDS, J. Remote sensing digital image analysis. [S.l.]: Springer, 1999. v. 3. 21 ADENIYI, P. O. Digital analysis of multitemporal landsat data for land-use/land-cover classification in a semi-arid area of nigeria. Photogrammetric engineering and remote sensing, v. 51, n. 11, p. 1761–1774, 1985. 22 MONARD, M. C.; BARANAUSKAS, J. A. Indução de regras e árvores de decisão. Sistemas Inteligentes-Fundamentos e Aplicações, sn, v. 1, p. 115–139, 2003. 23 BREIMAN, L. Random forests. Machine learning, Springer, v. 45, n. 1, p. 5–32, 2001. 24 GOLDSCHMIDT, R.; PASSOS, E. Data mining. [S.l.]: Gulf Professional Publishing, 2005. 25 HAYKIN, S. Redes neurais: princípios e prática. [S.l.]: Bookman Editora, 2007. 26 CORTES, C.; VAPNIK, V. Support-vector networks. Machine learning, Springer, v. 20, n. 3, p. 273–297, 1995. 27 RAUBER, T. W. Redes neurais artificiais. Universidade Federal do Espírito Santo, v. 29, 2005. 28 SILVA, L. Uma aplicação de árvores de decisão, redes neurais e knn para a identificação de modelos arma não sazonais e sazonais. Rio de Janeiro. 145p. Tese de Doutorado-Departamento de Engenharia Elétrica, Pontifícia Universidade Católica do Rio de Janeiro, 2005. 40 29 GREGÓRIO, R.; OLIVEIRA, P. R. Método proximal com fatoraçoes de schur para determinaçao da média riemanniana de matrizes simétricas definidas positivas. 2012. 30 SAKAI, T. Riemannian geometry, translations of mathematical monographs 149, amer-ican mathematical society, providence, ri, 1996. MR1390760 (97f: 53001). 31 CARMO, M. P. d. Geometria Riemanniana. [S.l.], 1988. 32 LEON, S. J. Álgebra Linear com Aplicações . [S.l.]: Grupo Gen-LTC, 2000. 33 MOAKHER, M. A differential geometric approach to the geometric mean of symmetric positive-definite matrices. SIAM Journal on Matrix Analysis and Applications, SIAM, v. 26, n. 3, p. 735–747, 2005. 34 GREGÓRIO, R.; OLIVEIRA, P. R. Título Algoritmo de Ponto no Cone das Matrizes Simétricas Semidefinidas Positivas e Um Método de Escalarização Proximal Log Quadrático para Programação Multiobjetivo. Tese (Doutorado) — COPPE/UFRJ, 2008. 35 ROTHAUS, O. S. Domains of positivity. In: SPRINGER. Abhandlungen aus dem Mathematischen Seminar der Universität Hamburg. [S.l.], 1960. v. 24, n. 1, p. 189–235. 36 NESTEROV, Y. E.; TODD, M. J. et al. On the riemannian geometry defined by self- concordant barriers and interior-point methods. Foundations of Computational Mathematics, Springer, v. 2, n. 4, p. 333–361, 2002. 37 HARANDI, M. T. et al. Sparse coding and dictionary learning for symmetric positive definite matrices: A kernel approach. In: SPRINGER. European Conference on Computer Vision. [S.l.], 2012. p. 216–229. 38 LI, P. et al. Log-euclidean kernels for sparse representation and dictionary learning. In: Proceedings of the IEEE international conference on computer vision. [S.l.: s.n.], 2013. p. 1601–1608. 39 CASEIRO, R. et al. Semi-intrinsic mean shift on riemannian manifolds. In: SPRINGER. European conference on computer vision. [S.l.], 2012. p. 342–355. 40 VEMULAPALLI, R.; PILLAI, J. K.; CHELLAPPA, R. Kernel learning for extrinsic classification of manifold features. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S.l.: s.n.], 2013. p. 1782–1789. 41 HUANG, Z. et al. Learning euclidean-to-riemannian metric for point-to-set classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S.l.: s.n.], 2014. p. 1677–1684. 42 HUANG, Z. et al. Log-euclidean metric learning on symmetric positive definite manifold with application to image set classification. In: PMLR. International conference on machine learning. [S.l.], 2015. p. 720–729. 43 DU, Y. et al. Classification of tumor epithelium and stroma by exploiting image features learned by deep convolutional neural networks. Annals of biomedical engineering, Springer, v. 46, n. 12, p. 1988–1999, 2018. 41 44 HEIDARI, M. et al. Prediction of breast cancer risk using a machine learning approach embedded with a locality preserving projection algorithm. Physics in Medicine & Biology, IOP Publishing, v. 63, n. 3, p. 035020, 2018. 45 HEIDARI, M. et al. Development and assessment of a new global mammographic image feature analysis scheme to predict likelihood of malignant cases. IEEE transactions on medical imaging, IEEE, v. 39, n. 4, p. 1235–1244, 2019. 46 OPBROEK, A. V. et al. Transfer learning improves supervised image segmentation across imaging protocols. IEEE transactions on medical imaging, IEEE, v. 34, n. 5, p. 1018–1030, 2014. 47 ZARGARI, A. et al. Prediction of chemotherapy response in ovarian cancer patients using a new clustered quantitative image marker. Physics in Medicine & Biology, IOP Publishing, v. 63, n. 15, p. 155020, 2018. 48 KHUZANI, A. Z.; HEIDARI, M.; SHARIATI, S. A. Covid-classifier: An automated machine learning model to assist in the diagnosis of covid-19 infection in chest x-ray images. Scientific Reports, Nature Publishing Group, v. 11, n. 1, p. 1–6, 2021. 49 OH, Y.; PARK, S.; YE, J. C. Deep learning covid-19 features on cxr using limited training data sets. IEEE transactions on medical imaging, IEEE, v. 39, n. 8, p. 2688–2700, 2020. 50 ROY, S. et al. Deep learning for classification and localization of covid-19 markers in point-of-care lung ultrasound. IEEE transactions on medical imaging, IEEE, v. 39, n. 8, p. 2676–2687, 2020. 51 WANG, S. et al. A fully automatic deep learning system for covid-19 diagnostic and prognostic analysis. European Respiratory Journal, Eur Respiratory Soc, v. 56, n. 2, 2020. 52 ZHANG, K. et al. Clinically applicable ai system for accurate diagnosis, quantitative measurements, and prognosis of covid-19 pneumonia using computed tomography. Cell, Elsevier, v. 181, n. 6, p. 1423–1433, 2020. 53 SONKA, M.; HLAVAC, V.; BOYLE, R. Image processing, analysis, and machine vision. [S.l.]: Cengage Learning, 2014.pt_BR
dc.subject.cnpqMatemáticapt_BR
dc.subject.cnpqMatemáticapt_BR
Appears in Collections:Mestrado em Modelagem Matemática e Computacional

Se for cadastrado no RIMA, poderá receber informações por email.
Se ainda não tem uma conta, cadastre-se aqui!

Files in This Item:
File Description SizeFormat 
2022 - Helainy Ignacio De Almeida Tôrres.pdf2.52 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.