Universiti Teknologi Malaysia Institutional Repository

A new approach for image segmentation and indexing using color,shape and location

Reza, Md. Parvez and Islam, Md. Rafiqul and Md. Sap, Mohd. Noor and Khatun, Fahmida and Kanti Mandal, Koushik (2004) A new approach for image segmentation and indexing using color,shape and location. Jurnal Teknologi Maklumat, 16 (2). pp. 55-73. ISSN 0128-3790

[img] PDF
362Kb

Abstract

Color is the most dominant and distinguishing visual feature for image. A compact color descriptor and an efficient indexing method for this descriptor are presented. The proposed descriptor is quite compact and is based on the observation that a small number of colors are sufficient to characterize the color information in an image region. The color feature extraction starts with color image segmentation. A modified color segmentation algorithm is proposed using reduced number of image pixels. Segmentation is performed based on its representative colors. Therefore no color clustering normally is required. For efficient search and retrieval an efficient indexing is presented. Hash indexing structure is used to index similar colored regions, which effectively improv~s the searching performance. Addition to color location and shape is also used to get the accurate image. Therefore a quick segmentation using a reduced number of colors and an efficient indexing scheme based on combined color, shape and location is presented.

Item Type:Article
Uncontrolled Keywords:image database, quantization, segmentation, hashing, eccentricity
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:Computer Science and Information System (Formerly known)
ID Code:3426
Deposited By: Mrs Rozilawati Dollah @ Md Zain
Deposited On:24 May 2007 05:18
Last Modified:30 Aug 2012 10:25

Repository Staff Only: item control page