Please use this identifier to cite or link to this item: https://rima.ufrrj.br/jspui/handle/20.500.14407/9396
Full metadata record
DC FieldValueLanguage
dc.contributor.authorAndrade, Bruno Geike de
dc.date.accessioned2023-12-21T18:38:25Z-
dc.date.available2023-12-21T18:38:25Z-
dc.date.issued2020-02-18
dc.identifier.citationANDRADE, Bruno Geike de. Visão computacional para identificação de espécies lenhosas em campo. 2020.104 f. Tese (Doutorado em Ciências Ambientais e Florestais) - Instituto de Florestas, Universidade Federal Rural do Rio de Janeiro, Seropédica, 2020.por
dc.identifier.urihttps://rima.ufrrj.br/jspui/handle/20.500.14407/9396-
dc.description.abstractA identificação anatômica de espécies florestais é uma importante ferramenta para controle e fiscalização do comércio de madeira, principalmente por possibilitar a proteção de espécies vulneráveis. O recente aumento das exigências do mercado internacional de madeira e a plena evolução de áreas tecnológicas como machine learning e machine vision têm incentivado o desenvolvimento de sistemas inteligentes e automáticos para identificação de espécies lenhosas a partir de imagens da madeira. Neste trabalho, buscou-se desenvolver e avaliar um sistema de visão computacional com uso de um smartphone para a aquisição de imagens de amostras de madeira polidas manualmente com facas. Três bancos de imagens foram construídos, o primeiro contendo 2.000 imagens de 21 espécies e o segundo contendo 30.200 imagens de 40 espécies. O terceiro, também com 40 espécies, foi formado com 32.271 imagens obtidas com amostras de madeira umedecidas superficialmente. Três descritores de imagens foram avaliados: grey level coocurrence matrix, local binary patterns e grey level run length matrix. Também foram avaliadas diferentes configurações de resolução e níveis de cinzas das imagens. Um total de 49 classificadores estatísticos foram desenvolvidos usando-se support vector machines e avaliados em validações cruzadas aninhadas. A grande maioria dos classificadores testados apresentaram taxas de acerto superior a 90%, local binary patterns apresentou desempenho superior aos demais descritores de imagem e o umedecimento das amostras não apresentou melhora no desempenho da classificação. O sistema proposto foi capaz de alcançar uma taxa de acerto de 99,36%, superando os resultados obtidos em todos os trabalhos da literatura consultada. A metodologia simples usada neste trabalho, associada à elevada taxa de acerto, torna evidente o potencial para a automatização da identificação de madeiras com sistema de visão computacional em condições de campopor
dc.description.sponsorshipCAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superiorpor
dc.formatapplication/pdf*
dc.languageporpor
dc.publisherUniversidade Federal Rural do Rio de Janeiropor
dc.rightsAcesso Abertopor
dc.subjectAnatomia da madeirapor
dc.subjectReconhecimento de espéciespor
dc.subjectMachine learningpor
dc.subjectWood anatomyeng
dc.subjectWood identificationeng
dc.subjectMachine learningpor
dc.titleVisão computacional para identificação de espécies lenhosas em campopor
dc.title.alternativeMachine vision for field-level wood identificationeng
dc.typeDissertaçãopor
dc.description.abstractOtherAnatomical identification of forestry species is an important tool for control and supervision of timber trade, mainly because it enables the protection of vulnerable species. The recent increase in the demands of the international timber market and the full evolution of technological areas such as machine learning and machine vision have encouraged the development of intelligent and automatic species identification systems based on wood images. In this work, we sought to develop and evaluate a computer vision system capable of identifying species with a smartphone to acquire images of manually polished samples with knives. Three image banks were built, the first containing 2,000 images of 21 species and the second containing 30,200 images of 40 species. The third, also with 40 species, was formed with 32,271 images obtained with superficial moistened wood samples. Three image descriptors were evaluated: gray level coocurrence matrix, local binary patterns and gray level run length matrix. Different resolution settings and gray levels of the images were also evaluated. A total of 49 statistical classifiers were developed using support vector machines and evaluated in nested cross validations. The great majority of the classifiers tested presented accuracies higher than 90%, local binary patterns performed better than the other image descriptors and the sample wetting did not improve the classification performance. The proposed system was able to reach 99.36% accuracy, surpassing the results obtained in all works of the consulted literature. The simple methodology used in this work, associated with this high accuracy, makes evident the potential for the automated identification of wood with machine vision system under field conditionseng
dc.contributor.advisor1Latorraca, João Vicente de Figueiredo
dc.contributor.advisor1ID284.741.551-34por
dc.contributor.advisor1IDhttps://orcid.org/0000-0002-5969-5199por
dc.contributor.advisor1Latteshttp://lattes.cnpq.br/9612404360795583por
dc.contributor.referee1Latorraca, João Vicente de Figueiredo
dc.contributor.referee1ID284.741.551-34por
dc.contributor.referee1IDhttps://orcid.org/0000-0002-5969-5199por
dc.contributor.referee1Latteshttp://lattes.cnpq.br/9612404360795583por
dc.contributor.referee2Costa, Anderson Gomide
dc.contributor.referee2IDhttps://orcid.org/0000-0003-0594-8514por
dc.contributor.referee2Latteshttp://lattes.cnpq.br/6959807888629144por
dc.contributor.referee3Mendonca, Bruno Araujo Furtado de
dc.contributor.referee3IDhttps://orcid.org/0000-0003-0288-0024por
dc.contributor.referee3Latteshttp://lattes.cnpq.br/8081324794152785por
dc.contributor.referee4Muniz, Graciela Ines Bolzon de
dc.contributor.referee4Latteshttp://lattes.cnpq.br/4038930548278283por
dc.contributor.referee5Moulin, Jordão Cabral
dc.contributor.referee5IDhttps://orcid.org/0000-0002-5543-3853por
dc.contributor.referee5Latteshttp://lattes.cnpq.br/3577181658928552por
dc.creator.ID058.745.617-57por
dc.creator.Latteshttp://lattes.cnpq.br/6940038437988975por
dc.publisher.countryBrasilpor
dc.publisher.departmentInstituto de Florestaspor
dc.publisher.initialsUFRRJpor
dc.publisher.programPrograma de Pós-Graduação em Ciências Ambientais e Florestaispor
dc.relation.referencesANDRADE, B. G. Discriminação entre carvões provenientes de madeira de Eucalyptus sp. e de espécies nativas por análise de imagens digitais. 2013. Universidade Federal de Viçosa, 2013. ANDRADE, B. G. et al. Potential of texture analysis for charcoal classification. Floresta e Ambiente, v. 26, n. 3, p. 10, 2019. ANGYALOSSY, V. et al. IAWA list of microscopic features for hardwood identification. IAWA Bulletin, v. 10, n. 3, p. 219–332, 1989. BACKES, A. R.; SÁ JUNIOR, J. J. M. Introdução à visão computacional usando MATLAB. Rio de Janeiro: Alta Books, 2016. 290 p. BISHOP, C. M. Pattern recognition and machine learning. New York: Springer, 2006. 681 p. BOTOSSO, P. Identificação macroscópica de madeiras. Colombo: Embrapa Florestas, 2011. 64 p. BURGUER, L. M.; RICHTER, H. G. Anatomia da madeira. São Paulo: Nobel, 1991. 154 p. DE MUÑIZ, G. I. B. et al. Aanálisis de la estructura anatómica de la madera y del carbón de dos especies de sapotaceae. Maderas: Ciencia y Tecnologia, v. 15, n. 3, p. 311–320, 2013. FAO. European Union and FAO to step up efforts to combat illegal timber trade. 2016. Disponível em: <http://www.fao.org/news/story/en/item/414279/icode/>. Acesso em: 6 set. 2019. GALLOWAY, M. M. Texture analysis using gray level run lengths. Computer Graphics and Image Processing, v. 4, n. 2, p. 172–179, 1975. Disponível em: <https://www.sciencedirect.com/science/article/pii/S0146664X75800086>. Acesso em: 11 out. 2019. GALVÃO, A. P. M.; JANKOWSKY, I. P. Secagem racional da madeira. São Paulo: Nobel, 1985. 111 p. 102 GEIRHOS, R. et al. Generalisation in humans and deep neural networks. In: Advances in Neural Information Processing Systems, Montréal. Anais... Montréal: 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), 2018. GONZALEZ, R. C.; WOODS, R. E.; EDDINS, S. . Digital image processing using matlab. Nova Jersey: Pearson Education, 2004. 344 p. HANSSEN, F. et al. Molecular identification of Fitzroya cupressoides, Sequoia sempervirens, and Thuja plicata wood using taxon-specific RDNA-ITS primers. IAWA Journal, v. 32, p. 273–284, 2011. HARALICK, R. M.; SHANMUGAM, K.; DINSTEIN, I. Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics, v. SMC-3, n. 6, p. 610–621, 1973. HASTIE, T.; TIBSHIRANI, R.; FRIEDMAN, J. The elements of statistical learning: data mining, inference, and prediction. New York: Springer, 2009. 763 p. HERMANSON, J. C.; WIEDENHOEFT, A. C. A brief review of machine vision in the context of automated wood identification systems. IAWA Journal, v. 32, n. 2, p. 233–250, 2011. HOARE, A. Tackling illegal logging and the related trade: what progress and where next? Disponível em: <https://www.chathamhouse.org/publication/tackling-illegal-logging-and-related-trade-what-progress-and-where-next>. Acesso em: 6 set. 2019. IBRAHIM, I. et al. Tree species recognition system based on macroscopic image analysis. Wood Science and Technology, v. 51, n. 2, p. 431–444, 2017. JAIN, A. K.; DUIN, R. P. W.; MAO, J. Statistical pattern recognition: a review. IEEE Transactions on Pattern Analysis and Machine Intelligence, v. 22, n. 1, p. 4–37, 2000. JAMES, G. et al. An introduction to statistical learning with applications in R. New York: Springer, 2014. 440 p. KHALID, M. et al. Design of an intelligent wood species recognition system. International Journal of Simulation Systems, Science & Technology, v. 9, n. 3, p. 11, 2008. KITE, G. C. et al. Dalnigrin, a neoflavonoid marker for the identification of Brazilian rosewood (Dalbergia Nigra) in CITES enforcement. Phytochemistry, v. 71, n. 10, p. 1122–1131, 2010. KOCH, G.; HAAG, V. Control of internationally traded timber - the role of macroscopic and microscopic wood identification againstcillegal logging. Journal of Forensic Research, v. 06, n. 06, p. 4, 2016. MAINIERI, C. Manual de identificaçõo das principais madeiras comerciais brasileiras. São Paulo: Instituto de Pesquisas Tecnologicas do Estado de São Paulo - IPT, 1983. 241 p.. MARTINS, J. G. et al. Forest species recognition based on dynamic classifier selection and dissimilarity feature vector representation. Machine Vision and Applications, v. 26, n. 2–3, p. 279–293, 2015. MARUYAMA, T. M. et al. Automatic classification of native wood charcoal. Ecological 103 Informatics, v. 46, p. 1–7, 2018. MATHWORKS. Image processing yoolbox TM reference R 2018a. Natick: The MathWorks, Inc., 2018. 2792 p. MATHWORKS. Statistics and machine learning toolboxTM user’s guide R2019b. Natick: The MathWorks, Inc., 2019. 9954 p. MICHEAL, A. A.; ANAND, T. A novel image classification method using texture feature descriptors. IOSR Journal of Engineering, v. 08, n. 5, p. 24–32, 2018. NÉMETH, R. et al. The effect of moisture content and drying temperature on the colour of two poplars and robinia wood. BioResources, v. 8, n. 2, p. 2074–2083, 2013. OJALA, T.; PIETIKÄINEN, M.; HARWOOD, D. A comparative study of texture measures with classification based on feature distributions. Pattern Recognition, v. 29, n. 1, p. 51–59, 1996. OLIVEIRA, A. A. et al. Identificação de madeiras utilizando a espectrometria no infravermelho próximo e redes neurais artificiais. Tema, v. 16, n. 2, p. 81, 2015. PANSHIN, A. J.; DE ZEEUW, C. Textbook of wood technology. New York: McGraw-Hill, 1964. 705 p. PAULA FILHO, P. L. et al. Forest species recognition using macroscopic images. Machine Vision and Applications, v. 25, n. 4, p. 1019–1031, 2014. PEDRINI, H.; SCHWARTZ, W. R. Analise de imagens digitais - principios, algoritmos e aplicacoes. São Paulo: Thomson Learning, 2008. 528 p. PETROU, M.; SEVILLA, P. G. Image Processing: Dealing with Texture. Chichester: John Wiley & Sons, 2006. QIN, X.; YANG, Y. H. Basic gray level aura matrices: theory and its application to texture synthesis. Proceedings of the IEEE International Conference on Computer Vision, v. I, n. 4, p. 128–135, 2005. RAVINDRAN, P. et al. Classification of CITES-listed and other neotropical Meliaceae wood images using convolutional neural networks. Plant Methods, v. 14, n. 1, p. 1–10, 2018. Disponível em: <https://doi.org/10.1186/s13007-018-0292-9>. Acesso em: 6 set. 2019. RUFFINATTO, F.; CRIVELLARO, A.; WIEDENHOEFT, A. C. Review of macroscopic features for hardwood and softwood identification and a proposal for a new character list. IAWA Journal, v. 36, n. 2, p. 208–241, 2015. SHALEV-SHWARTZ, S.; BEN-DAVID, S. Understanding machine learning: from theory to algorithms. New York: Cambridge University Press, 2014. SOLOMON, C.; BRECKON, T.; SOUZA, J. R. Fundamentos de processamento digital de imagens : uma abordagem prâtica com exemplos em Matlab. Rio de Janeiro: LTC, 2016. SPIEGEL, M. R.; LIPSCHUTZ, S.; SPELLMAN, D. Vector analysis (Schaum’s outlines). New York: McGraw Hill Education, 2009. 264 p. TANG, X. Texture information in run-length matrices. IEEE Transactions on Image 104 Processing, v. 7, n. 11, p. 1602–1609, 1998. TAYLOR, K. Pattern recognition and classification using Matlab. Scotts Valley: CreateSpace Publishing, 2017. 360 p. THEODORIDIS, S. et al. Introduction to pattern recognition: a Matlab approach. Burlington: Academic Press, 2010. THEODORIDIS, S.; KOUTROUMBAS, K. Pattern recognition. San Diego: Academic Press, 2009. TOU, J. Y.; TAY, Y. H.; LAU, P. Y. A comparative study for texture classification techniques on wood species recognition problem. 5th International Conference on Natural Computation, ICNC 2009, v. 5, p. 8–12, 2009. VAROQUAUX, G. et al. Assessing and tuning brain decoders: cross-validation, caveats, and guidelines. NeuroImage, v. 145, p. 166–179, 2017. Disponível em: <http://dx.doi.org/10.1016/j.neuroimage.2016.10.038>. Acesso em: 6 set. 2019. WÄLDCHEN, J.; MÄDER, P. Plant species identification using computer vision yechniques : a systematic literature review. Archives of Computational Methods in Engineering, v. 25, p. 507–543, 2018. WANG, H.; ZHANG, G.; QI, H. Wood recognition using image texture features. PLoS ONE, v. 8, n. 10, p. 1–12, 2013. WEBB, A. R. Statistical pattern recognition. Chichester: Wiley, 2002. 496 p. WEI, X. Gray level run length matrix toolbox, 2007. . Disponível em: <https://www.mathworks.com/matlabcentral/fileexchange/17482-gray-level-run-length-matrix-toolbox> Acesso em: 6 set. 2019. WHEELER, E. A.; BAAS, P. Wood identification - a review. IAWA Journal, v. 19, n. 3, p. 241–264, 1998. YUSOF, R.; KHALID, M.; ANIS, A. S. Application of kernel-genetic algorithm as nonlinear feature selection in tropical wood species recognition system. Computers and Electronics in Agriculture, v. 93, p. 68–77, 2013. Disponível em: <http://dx.doi.org/10.1016/j.compag.2013.01.007>. Acesso em: 6 set. 2019. ZAMRI, M. I. P. Bin et al. Statistical feature extraction method for wood species recognition system. International Journal of Computer and Information Engineering, v. 10, n. 3, p. 441–444, 2016.por
dc.subject.cnpqRecursos Florestais e Engenharia Florestalpor
dc.thumbnail.urlhttps://tede.ufrrj.br/retrieve/70875/2020%20-%20Bruno%20Geike%20de%20Andrade.pdf.jpg*
dc.originais.urihttps://tede.ufrrj.br/jspui/handle/jspui/6024
dc.originais.provenanceSubmitted by Celso Magalhaes (celsomagalhaes@ufrrj.br) on 2022-09-29T13:55:00Z No. of bitstreams: 1 2020 - Bruno Geike de Andrade.pdf: 5366325 bytes, checksum: c0bb4228f0aaf0370b2189ea6a6f19b8 (MD5)eng
dc.originais.provenanceMade available in DSpace on 2022-09-29T13:55:00Z (GMT). No. of bitstreams: 1 2020 - Bruno Geike de Andrade.pdf: 5366325 bytes, checksum: c0bb4228f0aaf0370b2189ea6a6f19b8 (MD5) Previous issue date: 2020-02-18eng
Appears in Collections:Doutorado em Ciências Ambientais e Florestais

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 
2020 - Bruno Geike de Andrade.pdf2020 - Bruno Geike de Andrade5.24 MBAdobe PDFThumbnail
View/Open


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