A modified hybrid DCT-SVD image-coding system

Wongsawat, Yodchanan; Ochoa, Humberto; Rao, K. R.

doi:10.1109/tencon.2004.1414425

Cited by 4 publications

(4 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As it can be seen from the results choosing a constant value of q is impractical because the amount of information is different from block to block. Figure (5) shows that the MSE method is better than a percentage of the discarded singular value method for selecting the retained singular values.…”

Section: Resultsmentioning

confidence: 99%

“…Many authors used the SVD combined with other transforms to achieve a good compression results. In [3,5], the image is divided into blocks of 8*8 pixels, and the SVD or DCT transform is applied to each block depends on the standard deviation of that block. Other works [6,7] combine the SVD with the DWT, the image is partitioned into tiles of 64*64 pixels, the Average Standard Deviation (ASTD) is computed for each tile depends on the standard deviations of 8*8 subblocks of the tile.…”

Section: Related Workmentioning

confidence: 99%

“…Using a specified Mean Square Error (MSE) for the image or (sub_image) to determine the value of q [7]. A MSE can be computed as in [5].…”

Section: Svd Transform and Image Compressionmentioning

confidence: 99%

“…In addition, the compression obtained based on SVD is modest because the eigenvalues and its associated eigenvectors must be coded [3]. The DCT, DWT, and SVD transforms are combined in many works [1,3,5,6,7] to achieve better compression results. In most of these works, the image is divided into blocks and a proper transform is applied to each block depending on a predefined criteria.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Comparison of the Methods used for Selecting Singular values in Image Compression using SVD

Khalid¹

2018

IJCA

View full text Add to dashboard Cite

In this paper, an image compression using singular value decomposition (SVD) transform is presented. The SVD decomposes the image into two eigenvector matrices and a one singular value diagonal matrix. The compression is achieved by selecting some singular values and their associated eigenvectors. The proper selection of the retained singular values is the critical issues in image compression based SVD transform. The SVD transform is applied to the entire image, and also the image is divided into blocks with the SVD applied to each block. The objective of this paper is to study and discuss the methods used to select the singular values that achieve an acceptable image quality with a reduced size.

show abstract

Section: Resultsmentioning

confidence: 99%