Paper Details: Downloads: 668
Serial Number: P1121051397
Title: Sketch-based Shape Retrieval relying on BSR Technique
Authors: N. D. Salih and Rosli Besar and F. S. Abas
Abstract: Due to the expansion of multimedia information, the challenging problem of designing techniques that support effectively searching through large volume image databases has arisen. In response to such a challenge, Content-based image retrieval (CBIR) has become a major area of research and received increased attention. Representation and description of shapes is one of the major problems in CBIR. This paper present an a novel method for shape representation and description named block-based shape representation (BSR), which is capable of extracting reliable information of the object outline in a concise manner.BSR is mainly designed to retrieve man-made objects based on their shape using the user’s sketch as a query input. This technique has been implemented and evaluated in order to analyze its accuracy and efficiency. Based on the experimental results, we urge that the proposed BSR is a compact and reliable shape representation method and fast enough for use in real-time retrieval systems.
Keywords: CBIR, Shape representation, Shape descriptor
Journal/Conference: International Journal of Artificial Intelligence and Machine Learning
Volume: 11
Issue: 1
Submission Date: 12/15/2010 12:00:00 AM
Review Date: 5/6/2011 12:00:00 AM
Publishing Date: 6/11/2011 12:00:00 AM
Article Downloads: 668
Download:

Facebook