Aplikasi Image Retrieval dengan Histogram Warna dan Multi-scale GLCM

Arwin Halim, Hardy Hardy, Mytosin Mytosin


Content-based image retrieval is an image search techniques from large image database by analyzing features of the image. Image feature can be color, texture, shape, and others. This study uses color and texture features when searching image. Color histogram is used to extract color features with quantization approach to HSV. Texture features in image obtained from the calculation of Gray-Level Co-occurrence Matrix (GLCM) and multi-scale GLCM. Multi-scale GLCM using Gaussian smoothing to reduce noise in the image and considering multiple scale from an image. Image search results obtained from the comparison of the features of color and texture in database using Euclidean distance. The results show an image search on Wang database using color histogram and multi-scale GLCM obtain higher precision value than just taking one of the method or combinations of color histogram and GLCM.


content-based image retrieval, color histogram, multi-scale GLCM

Full Text: PDF


  • There are currently no refbacks.

Lembaga Penelitian & Pengabdian pada Masyarakat (LPPM)
Universitas Mikroskil
Jl. Thamrin No. 124 Medan - 20212
Gedung A. 07.L2
Telp. 061-4573767
Email: publication@mikroskil.ac.id

Creative Commons License
The JSM site and its metadata are licensed under CC BY-NC-ND