Sparse orthonormal transforms for image compression

We propose a block-based transform optimization and associated image compression technique that exploits regularity along directional image singularities. Unlike established work, directionality comes about as a byproduct of the proposed optimization rather than a built in constraint. Our work class...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2008 15th IEEE International Conference on Image Processing s. 149 - 152
Hlavní autoři: Sezer, O.G., Harmanci, O., Guleryuz, O.G.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.10.2008
Témata:
Codecs
Constraint optimization
Cost function
Design optimization
directional transforms
Image coding
Image processing
Rate-distortion
Signal processing
sparse lapped transforms
Sparse orthonormal transforms
Transform coding
Wavelet transforms
ISBN:9781424417650, 1424417651
ISSN:1522-4880
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:We propose a block-based transform optimization and associated image compression technique that exploits regularity along directional image singularities. Unlike established work, directionality comes about as a byproduct of the proposed optimization rather than a built in constraint. Our work classifies image blocks and uses transforms that are optimal for each class, thereby decomposing image information into classification and transform coefficient information. The transforms are optimized using a set of training images. Our algebraic framework allows straightforward extension to non-block transforms, allowing us to also design sparse lapped transforms that exploit geometric regularity. We use an EZW/SPIHT like entropy coder to encode the transform coefficients to show that our block and lapped designs have competitive rate-distortion performance. Our work can be seen as nonlinear approximation optimized transform coding of images subject to structural constraints on transform basis functions.
ISBN:9781424417650
1424417651
ISSN:1522-4880
DOI:10.1109/ICIP.2008.4711713