Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Sains dan Teknologi
Zubaidah, Rima Puti
Unknown Affiliation
Aerospace Engineering Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Earth & Planetary Sciences Electrical & Electronics Engineering Energy Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mathematics Mechanical Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Penerapan K-Means Clustering untuk Klasifikasi Citra Aksesoris Ekstraksi Warna dan Tekstur GLCM Zubaidah, Rima Puti; Ramadhanu, Agung
Jurnal Sains dan Teknologi (JSIT) Vol. 5 No. 3 (2025): September-Desember
Publisher : CV. Information Technology Training Center - Indonesia (ITTC)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47233/jsit.v5i3.3633

Abstract

The main problem in accessory image recognition lies in the similarity of physical shapes among objects such as bracelets, necklaces, and earrings, which often causes difficulties in the automatic classification process. This study aims to develop an accessory image classification system capable of accurately grouping objects based on a combination of color and texture features using the K-Means Clustering algorithm. The method used includes several preprocessing stages such as resizing images to ensure uniform dimensions and normalizing pixel values to achieve consistent data scales. Color features were extracted using RGB and HSV histograms to represent color variations, while texture features were obtained through the Gray Level Co-occurrence Matrix (GLCM) method with four parameters: contrast, correlation, energy, and homogeneity. All extracted features were then combined and analyzed using the K-Means algorithm with k=3, corresponding to the number of accessory categories. The results show that combining color and texture features produces a more optimal cluster separation compared to using single-feature extraction. The K-Means algorithm successfully grouped accessory images according to their respective categories with high consistency. These findings have potential applications in digital catalog management systems and product recommendation systems on e-commerce platforms.
Co-Authors Ramadhanu, Agung