Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent

Detalhes bibliográficos
Autor(a) principal: Cassiano, Fernando Ribeiro
Data de Publicação: 2022
Outros Autores: Sáfadi, Thelma, Guimarães, Paulo Henrique Sales
Tipo de documento: Artigo
Idioma: eng
Título da fonte: Research, Society and Development
Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/36211
Resumo: Modernization is present in all fields of knowledge. The wavelet transform and the Hurst exponent are tools that have fundamental importance in many of these advances. In the present study, the wavelet decomposition technique was combined with the Hurst exponent calculation to analyze X-ray images of seeds and thus classify them as full, slightly damaged or damaged. To calculate the Hurst exponent the mean and median were used as measurements of position. A support vector machine was used to validate the proposed method. For the full, damaged and slightly damaged seed groups, the average accuracy of the method, using the mean as measure position, was 74.5%, and using the median was 57.05%. For the full and damaged seed groups, the average accuracy using the mean was 99.76%, and using the median was 80.93%. For the slightly damaged and damaged seed groups, the average accuracy, using the mean as measure of position, was 99.26%, and the median was 76.22%. When analyzing seeds with slight damage, we observed a decrease in accuracy because the classification of the X-rays was subjective. Therefore, for the image database used in this study, the proposed methodology is efficient for automatic classification.
id UNIFEI_c6b89bcedecd7e7a29e7f75343a993e3
oai_identifier_str oai:ojs.pkp.sfu.ca:article/36211
network_acronym_str UNIFEI
network_name_str Research, Society and Development
repository_id_str
spelling Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent Exponente direccional de Hurst en el análisis de similitud de imágenes de semillasExpoente direcional de Hurst na análise de similaridade de imagens de sementes AutossimilaridadeClassificação de imagensOndaletas.Auto-similitudClasificación de imágenesOndiculas.AutossimilarityImage ClassificationWavelets.Modernization is present in all fields of knowledge. The wavelet transform and the Hurst exponent are tools that have fundamental importance in many of these advances. In the present study, the wavelet decomposition technique was combined with the Hurst exponent calculation to analyze X-ray images of seeds and thus classify them as full, slightly damaged or damaged. To calculate the Hurst exponent the mean and median were used as measurements of position. A support vector machine was used to validate the proposed method. For the full, damaged and slightly damaged seed groups, the average accuracy of the method, using the mean as measure position, was 74.5%, and using the median was 57.05%. For the full and damaged seed groups, the average accuracy using the mean was 99.76%, and using the median was 80.93%. For the slightly damaged and damaged seed groups, the average accuracy, using the mean as measure of position, was 99.26%, and the median was 76.22%. When analyzing seeds with slight damage, we observed a decrease in accuracy because the classification of the X-rays was subjective. Therefore, for the image database used in this study, the proposed methodology is efficient for automatic classification.La modernización está presente en todos los campos del conocimiento. Técnicas más sofisticadas y aparatos más modernos están surgiendo constantemente. La descomposición en ondiculas es una herramienta que tiene una importancia fundamental en muchos de estos avances. En lo que se refiere al análisis de imágenes, esta herramienta ha contribuido para creación de nuevas técnicas diferentes, como para la reconstrucción, comprensión y eliminación de ruidos, entre otros. Otra herramienta que auxilia en el análisis de imágenes es el exponente de Hurst, que mide cuanto tiene de auto-similitud una imagen, de forma que se capte información sobre las características de la imagen que a simple vista no seria posible. Con ello, el objetivo de este trabajo será combinar la técnica de descomposición en ondiculas con el cálculo del exponente de Hurst para analizar imágenes de semillas y así poder clasificarlas en llenas, levemente dañadas o dañadas. En el cálculo del exponente de Hurst serán usadas como medida de localización la media y la mediana. Un modelo de Máquinas de Vectores de Soporte será usado para la validación del método propuesto. Para el grupo de todas las semillas la precisión media del método, utilizando la media, fue de 74,5% y, con la mediana fue de 57,05%. Utilizando el grupo de semillas llenas y dañadas, la tasa media de precisión, con la media como medida de posición, fue de 99,76% y, con la mediana fue de 80,93%. En el grupo que contiene semillas levemente dañadas y dañadas la tasa media de precisión, usando la media como medida de posición, fue de 99,26% y, con la mediana fue de 76,22%.A modernização está presente em todos os campos do conhecimento. Técnicas mais sofisticadas e aparelhos mais modernos têm surgido com frequência. A decomposição em ondaletas é uma ferramenta que tem fundamental importância em muitos desses avanços. No que se refere à análise de imagens, essa ferramenta tem cooperado para criação de diversas novas técnicas, seja para reconstrução, compressão, eliminação de ruído, dentre outros. Outra ferramenta que auxilia na análise de imagens é o expoente de Hurst, que mede o quão autossimilar uma imagem é, de forma que se capte informação sobre características da imagem que a olho nu não seria possível. Com isso, o objetivo deste trabalho será combinar a técnica de decomposição em ondaletas com o cálculo do expoente de Hurst para analisar imagens de sementes e assim classificá-las em cheias, levemente danificadas ou danificadas. No cálculo do expoente de Hurst serão usadas como medidas de localização a média e a mediana. Um suport vector machine foi usado para validação do método proposto. Para o grupo de todas as sementes a acurácia média do método, utilizando a média, foi de 74,5% e, para a mediana foi de 57,05%. Utilizando o grupo de sementes cheias e danificadas a taxa média de acurácia, com a média como medida de posição, foi de 99,76% e, com a mediana foi de 80,93%. No grupo contendo sementes levemente danificadas e danificadas a taxa média de acurácia, usando a média como medida de posição, foi de 99,26% e, com a mediana foi de 76,22%.Research, Society and Development2022-10-27info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://rsdjournal.org/index.php/rsd/article/view/3621110.33448/rsd-v11i14.36211Research, Society and Development; Vol. 11 No. 14; e297111436211Research, Society and Development; Vol. 11 Núm. 14; e297111436211Research, Society and Development; v. 11 n. 14; e2971114362112525-3409reponame:Research, Society and Developmentinstname:Universidade Federal de Itajubá (UNIFEI)instacron:UNIFEIenghttps://rsdjournal.org/index.php/rsd/article/view/36211/30359Copyright (c) 2022 Fernando Ribeiro Cassiano; Thelma Sáfadi; Paulo Henrique Sales Guimarãeshttps://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessCassiano, Fernando Ribeiro Sáfadi, Thelma Guimarães, Paulo Henrique Sales 2022-11-08T13:36:27Zoai:ojs.pkp.sfu.ca:article/36211Revistahttps://rsdjournal.org/index.php/rsd/indexPUBhttps://rsdjournal.org/index.php/rsd/oairsd.articles@gmail.com2525-34092525-3409opendoar:2024-01-17T09:50:50.750957Research, Society and Development - Universidade Federal de Itajubá (UNIFEI)false
dc.title.none.fl_str_mv Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
Exponente direccional de Hurst en el análisis de similitud de imágenes de semillas
Expoente direcional de Hurst na análise de similaridade de imagens de sementes
title Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
spellingShingle Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
Cassiano, Fernando Ribeiro
Autossimilaridade
Classificação de imagens
Ondaletas.
Auto-similitud
Clasificación de imágenes
Ondiculas.
Autossimilarity
Image Classification
Wavelets.
title_short Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
title_full Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
title_fullStr Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
title_full_unstemmed Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
title_sort Wavelets in the analysis of seed image similarity: an approach using the Hurst directional exponent
author Cassiano, Fernando Ribeiro
author_facet Cassiano, Fernando Ribeiro
Sáfadi, Thelma
Guimarães, Paulo Henrique Sales
author_role author
author2 Sáfadi, Thelma
Guimarães, Paulo Henrique Sales
author2_role author
author
dc.contributor.author.fl_str_mv Cassiano, Fernando Ribeiro
Sáfadi, Thelma
Guimarães, Paulo Henrique Sales
dc.subject.por.fl_str_mv Autossimilaridade
Classificação de imagens
Ondaletas.
Auto-similitud
Clasificación de imágenes
Ondiculas.
Autossimilarity
Image Classification
Wavelets.
topic Autossimilaridade
Classificação de imagens
Ondaletas.
Auto-similitud
Clasificación de imágenes
Ondiculas.
Autossimilarity
Image Classification
Wavelets.
description Modernization is present in all fields of knowledge. The wavelet transform and the Hurst exponent are tools that have fundamental importance in many of these advances. In the present study, the wavelet decomposition technique was combined with the Hurst exponent calculation to analyze X-ray images of seeds and thus classify them as full, slightly damaged or damaged. To calculate the Hurst exponent the mean and median were used as measurements of position. A support vector machine was used to validate the proposed method. For the full, damaged and slightly damaged seed groups, the average accuracy of the method, using the mean as measure position, was 74.5%, and using the median was 57.05%. For the full and damaged seed groups, the average accuracy using the mean was 99.76%, and using the median was 80.93%. For the slightly damaged and damaged seed groups, the average accuracy, using the mean as measure of position, was 99.26%, and the median was 76.22%. When analyzing seeds with slight damage, we observed a decrease in accuracy because the classification of the X-rays was subjective. Therefore, for the image database used in this study, the proposed methodology is efficient for automatic classification.
publishDate 2022
dc.date.none.fl_str_mv 2022-10-27
dc.type.driver.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.uri.fl_str_mv https://rsdjournal.org/index.php/rsd/article/view/36211
10.33448/rsd-v11i14.36211
url https://rsdjournal.org/index.php/rsd/article/view/36211
identifier_str_mv 10.33448/rsd-v11i14.36211
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv https://rsdjournal.org/index.php/rsd/article/view/36211/30359
dc.rights.driver.fl_str_mv Copyright (c) 2022 Fernando Ribeiro Cassiano; Thelma Sáfadi; Paulo Henrique Sales Guimarães
https://creativecommons.org/licenses/by/4.0
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Copyright (c) 2022 Fernando Ribeiro Cassiano; Thelma Sáfadi; Paulo Henrique Sales Guimarães
https://creativecommons.org/licenses/by/4.0
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Research, Society and Development
publisher.none.fl_str_mv Research, Society and Development
dc.source.none.fl_str_mv Research, Society and Development; Vol. 11 No. 14; e297111436211
Research, Society and Development; Vol. 11 Núm. 14; e297111436211
Research, Society and Development; v. 11 n. 14; e297111436211
2525-3409
reponame:Research, Society and Development
instname:Universidade Federal de Itajubá (UNIFEI)
instacron:UNIFEI
instname_str Universidade Federal de Itajubá (UNIFEI)
instacron_str UNIFEI
institution UNIFEI
reponame_str Research, Society and Development
collection Research, Society and Development
repository.name.fl_str_mv Research, Society and Development - Universidade Federal de Itajubá (UNIFEI)
repository.mail.fl_str_mv rsd.articles@gmail.com
_version_ 1797052726599745536