Telecommunications and Radio Engineering

ISSN for PRINT: 0040-2508

For Online Access

Best Paper Award Selection - Editorial Board Site

2003, Volume59

I. V. Barishev
National Aerospace University (Kharkov Aviation Institute), 17, Chkalov St., Kharkov, 61070, Ukraine

M. L. Uss
National Aerospace University (Kharkov Aviation Institute), 17, Chkalov St., Kharkov, 61070, Ukraine

Filtering is one of the principal tasks of image processing. This problem is especially complicated due to the fact that images are very inhomogeneous in their structure and can be corrupted by additive, multiplicative or pulse interferences [1,2]. The algorithm of filtering has to solve two contradictory challenges - to suppress noise and to save boundaries and minor details in the picture. By now plenty of algorithms for filtering [1,2] have been developed, although improvement of the quality of saving boundaries and minor details still remains a crucial task for the present time.
Image filtering presumes a choice of a certain image model. A properly chosen model allows to describe the image with the prescribed accuracy, but providing for less amount of information. On the basis of the above model it is possible to carry out both compression and filtering of images, at that, the quality of both compression and filtering are usually rigidly bound. Presently, fractal compression methods [3], using self-similarity of the image at various scaling, are among the best compression algorithms, still very little attention is paid to the development of the methods of filtering based on this idea in special literature [4]. Therefore, an attempt to apply the ideas of fractal compression for image filtering has been undertaken in this paper.

DOI: 10.1615/TelecomRadEng.v59.i12.150 Download article, 8 pages

Article price - $35.00

<< Previous article   Next article >>