Abstract

When a great deal is known about the regularities in the files to be compressed, e.g., in the compression of image, audio, photo and video data, it is possible to match different compression algorithms with different parts of the data in such a way to achieve the maximum compression ratio, thereby providing significant reductions in file sizes. These algorithms are difficult to derive and are often the result of many person years of effort. Nonetheless, these are so successful that effectively some types of data are regularly saved, kept and transferred, only in compressed form, to be decompressed automatically and transparently to the user only when loaded into applications that make use of them. Our proposed approach is the combination of several approaches to make the compression better than the previous used approach. We first apply two Levels Discrete Wavelet Transform and then apply Walsh-Wavelet Transform on each 8x8 block of the low-frequency sub-band, then Split all DC values form each transformed block 8x8. Finally we perform compression by using arithmetic coding. In our algorithm we provide the basis of accepting images from the database. We concentrate the type of the wavelet we use like db1, db2, db3 etc. Then we use the quantization factor which is CF1 and CF2 in our case. We use the value as 0.05 and 0.2 as the quantization factor. After matlab simulation we can find our results suitable than the previous work.

Keyword

Walsh-Wavelet Transform, Image Compression, db1, db2, Quantization .

Cite this article

Refference