Sunflower Image Classification Using Multiclass Support Vector Machine
Based on Histogram Characteristics
Dublin Core
Title
Sunflower Image Classification Using Multiclass Support Vector Machine
Based on Histogram Characteristics
Based on Histogram Characteristics
Subject
histogram characteristics; first order feature extraction; support vector machine; multiclass SVM
Description
Sunflower is an important commodity in agriculture, besides being used as an ornamental plant, sunflower is an oil-producing
plant and a source of industrial materials. In Indonesia, sunflower productivity is considered less than optimal, because
knowledge and information about sunflowers are still lacking. Therefore, information is needed that can be used as an extension
of knowledge about sunflowers itself, especially in Indonesia, which is a tropical region which is an area suitable for the growth
of sunflowers. Sunflowers can actually be identified based on recognizable traits. However, the similar shape makes it difficult
for some people to distinguish the types of sunflowers. This study aims to classify sunflower images using a first-order feature
extraction algorithm using the characteristics of mean, skewness, variance, kurtosis, and entropy which are then used as input
to the Multiclass SVM identification algorithm. Data points are mapped to dimensionless space using a Multiclass SVM to
produce hyperplane-linear separation between each class. Based on the results of testing the accuracy of the model is able to
perform classification with an average accuracy of 79%. These results show that the developed model can classify well
plant and a source of industrial materials. In Indonesia, sunflower productivity is considered less than optimal, because
knowledge and information about sunflowers are still lacking. Therefore, information is needed that can be used as an extension
of knowledge about sunflowers itself, especially in Indonesia, which is a tropical region which is an area suitable for the growth
of sunflowers. Sunflowers can actually be identified based on recognizable traits. However, the similar shape makes it difficult
for some people to distinguish the types of sunflowers. This study aims to classify sunflower images using a first-order feature
extraction algorithm using the characteristics of mean, skewness, variance, kurtosis, and entropy which are then used as input
to the Multiclass SVM identification algorithm. Data points are mapped to dimensionless space using a Multiclass SVM to
produce hyperplane-linear separation between each class. Based on the results of testing the accuracy of the model is able to
perform classification with an average accuracy of 79%. These results show that the developed model can classify well
Creator
Rini Nuraini1
, Rachmat Destriana2
, Desi Nurnaningsih3
, Yeni Daniarti4
, Allan Desi Alexander5
, Rachmat Destriana2
, Desi Nurnaningsih3
, Yeni Daniarti4
, Allan Desi Alexander5
Publisher
Universitas Nasional
Date
03-02-2023
Contributor
Fajar bagus w
Format
PDF
Language
Indonesia
Type
Text
Files
Collection
Citation
Rini Nuraini1
, Rachmat Destriana2
, Desi Nurnaningsih3
, Yeni Daniarti4
, Allan Desi Alexander5, “Sunflower Image Classification Using Multiclass Support Vector Machine
Based on Histogram Characteristics,” Repository Horizon University Indonesia, accessed June 8, 2025, https://repository.horizon.ac.id/items/show/9349.
Based on Histogram Characteristics,” Repository Horizon University Indonesia, accessed June 8, 2025, https://repository.horizon.ac.id/items/show/9349.