Open AccessImproved Image Search and Retrieval based on Dominant Colors
Author(s) -
S. Karthik,
Srikanta Murthy,
Uma Kameswari Chembrolu,
Vidya Sridhar
Publication year - 2011
Publication title -
international journal of computer science and informatics
Language(s) - English
Resource type - Journals
ISSN - 2231-5292
DOI - 10.47893/ijcsi.2011.1023
Subject(s) - computer science , image retrieval , histogram , information retrieval , set (abstract data type) , content based image retrieval , visual word , similarity (geometry) , simplicity , artificial intelligence , image (mathematics) , pattern recognition (psychology) , philosophy , epistemology , programming language
Growth and development of multimedia technology has led to an exponential increase in visual information. Where traditional keyword based information retrieval techniques just do not meet the users demand, CBIR hopes to prevail. CBIR refers to the retrieval of images from a database using information derived from the images themselves rather than solely from accompanying text indices. In this paper we describe an approach of content based retrieval of images from a database, based on dominant colors in the foreground and background of the image. These dominant colors along with histogram and some statistical features form a substantial set to determine the overall similarity between the images. This technique is tested on Simplicity test database and promising results are observed.