Paper Details: Downloads: 693
Serial Number: P1151215166
Title: Fast Encoding Technique of Transformed VQ for Effective CBIR
Authors: Vidya Khapli1 and Anjali Bhalchandra2
Abstract: In this paper, a design of fast vector quantization (VQ) algorithm for coding of images in the transform domain is presented. To speed up the codebook design process, the feature of transform coding is combined with vector quantization. The transformed training set is obtained with wavelet transform with reduced computational complexity. The method utilizes mean-squared error and energy based selection schemes for good clustering of data vectors in the training space for encoding of Transformed Vector Quantization (TVQ). It is highly computationally efficient. Simulation results are presented to show the superior performance of the proposed method in terms of accuracy and retrieval time as compared to the standard Linde-Buzo-Gray algorithm for codebook design.
Keywords: Content based image retrieval, vector quantization, codebook, wavelet transform
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 12
Issue: 2
Submission Date: 4/9/2012 12:00:00 AM
Review Date: 5/10/2012 12:49:17 PM
Publishing Date: 5/17/2012 2:57:47 AM
Article Downloads: 693
Download:

Facebook