Title :
Performance Analysis of Image Compression Using Enhanced Vector Quantizer Designed with Self Organizing Feature Maps: The Quality Perspective
Author :
Dandawate, Y.H. ; Joshi, M.A.
Author_Institution :
Vishwakarma Inst. of Inf. Technol., Pune
Abstract :
This paper presents a novel approach for designing an enhanced vector quantizer (VQ) used for compression of images. The main aim of this paper is to propose a method for efficient VQ design, which will be well applicable to all kinds of images including medical images. For the performance evaluation of images, different quality measures along with conventionally used PSNR are used. The results with the conventional method for VQ design using self organized feature maps and our proposed are discussed and compared using quality measures such as structural content, Image fidelity, normalized correlation coefficient and a newly developed by researchers, mean structural similarity index (MSSIM) and universal quality index (UQI).
Keywords :
image coding; self-organising feature maps; vector quantisation; Image fidelity; enhanced vector quantizer; image compression; mean structural similarity index; medical images; normalized correlation coefficient; performance analysis; self organizing feature maps; structural content; universal quality index; Biomedical imaging; Color; Computational intelligence; Image coding; Image reconstruction; Image storage; Neural networks; Performance analysis; Self organizing feature maps; Vector quantization;
Conference_Titel :
Conference on Computational Intelligence and Multimedia Applications, 2007. International Conference on
Conference_Location :
Sivakasi, Tamil Nadu
Print_ISBN :
0-7695-3050-8
DOI :
10.1109/ICCIMA.2007.223
