DATA COMPRESSION BY USING SYMMETRICAL COMPONENT TRANSFORM

This investigation concerns the effectiveness of image data compression for digital signals using a new transform method termed the symmetrical com-ponent transform. This transform is based on a previously described matrix expansion of a spatially discrete image termed the symmetrical component expansion. Details of the symmetrical component expansion are provided and its relationship to the symmetrical component transform established. It is shown that the symmetrical component transform belong to the class of well known unitary transforms so that previous data compression results based on unitary transforms can be utilized. Data compression results for the symmetrical component transform are presented for several images with varying detail in the images. Results based on transforming the whole image as well segmenting the image into sub-images or blocks to reduce the computational requirements are described. Methods of alleviating the effects of block boundaries are also considered. These results are compared with the discrete cosine transform, the best known amongst all unitary transforms. The results show that for typical scenes the discrete cosine transform has better data compression capabilities than the symmetrical component transform, but the symmetrical component transform is almost twice as fast as the discrete cosine transform.

Name thesis:degree_name

Master of Science (M.Sc.)

Level thesis:degree_level

Masters

Grantor

University of Saskatchewan

Year dc:date.issued

1988

Author dc:creator

• Chen, Chun

Advisor dc:contributor.advisor

• Wacker, A. G.

Handle dc:identifier.uri

https://hdl.handle.net/10388/15587

OAI identifier oai:identifier

oai:harvest.usask.ca:10388/15587