| | |
Summary: SPARSE ORTHONORMAL TRANSFORMS FOR IMAGE COMPRESSION
Osman G. Sezer
, Oztan Harmanci
, Onur G. Guleryuz
Center for Signal and Image Processing, Georgia Institute of Technology, Atlanta, GA 30308, USA
DoCoMo Comm. Laboratories USA Inc., Palo Alto, CA 94304, USA
osman@ece.gatech.edu, {oharmanci, guleryuz}@docomolabs-usa.com
ABSTRACT
We propose a block-based transform optimization and asso-
ciated image compression technique that exploits regularity
along directional image singularities. Unlike established
work, directionality comes about as a byproduct of the pro-
posed optimization rather than a built in constraint. Our
work classifies image blocks and uses transforms that are
optimal for each class, thereby decomposing image informa-
tion into classification and transform coefficient information.
The transforms are optimized using a set of training images.
Our algebraic framework allows straightforward extension
|
