STUDI KOMPARASI KLASIFIKASI POLA TEKSTUR CITRA DIGITAL MENGGUNAKAN METODE K-MEANS DAN NAÏVE BAYES

Auliasari, Karina and Kertaningtyas, Mariza (2018) STUDI KOMPARASI KLASIFIKASI POLA TEKSTUR CITRA DIGITAL MENGGUNAKAN METODE K-MEANS DAN NAÏVE BAYES. Jurnal Informatika, 18 (2).

	Text Paper 9 Bu Karina Informatika.pdf Download (456kB)
	Text Cek Similarity Paper 9 Bu Karina Informatika.pdf Download (2MB)
	Text (Pereview) Peer Review Paper 9 Bu Karina Informatika.pdf Download (1MB)

Abstract

In this research was doing some performance testing using the k-means and naïve bayes method in classifying two types of image data sets with different texture patterns. The data set tested is the image data set of batik patterns and the brodatz pattern, feature of the image pattern used in this study is contrast and energy that obtained using the gray level co-assurance matrix (GLCM) method. The testing and analysis results show that the set of brodatz pattern image data has better prediction accuracy than the batik pattern image data set with a difference in value of 1.52%. For time parameters in generating contrast and energy feature values, batik pattern image data sets are faster to generate when compared to brodatz pattern image data sets with a time difference of 27.8 milliseconds. Similar results also occur in testing based on prediction time parameters, where the prediction time of batik pattern image data is faster than the brodatz pattern image data set with a time difference of 30.6 milliseconds. From testing using time parameters, it can be concluded that the set of brodatz pattern image data takes longer because the pattern of texture is not uniform, namely in one image there is a smooth and rough pattern because the image is an image with natural texture, different from the batik pattern image that has uniform repetition pattern so that the texture is more regular.

Item Type:	Article
Uncontrolled Keywords:	Texture, Image, Gray Level Co-Occurance Matrix, K-Means, Naïve Bayes
Subjects:	Engineering > Industrial Engineering
Divisions:	Fakultas Teknologi Industri > Teknik Industri S1
Depositing User:	Mr Sayekti Aditya Endra
Date Deposited:	24 Oct 2019 03:45
Last Modified:	26 Jun 2020 03:11
URI:	http://eprints.itn.ac.id/id/eprint/4401

Actions (login required)

View Item