Title :
Adaptive cascade architectures for image compression
Author :
Charalampidis, D.
Author_Institution :
Electr. Eng. Dept., Univ. of New Orleans, LA, USA
Abstract :
A novel image compression technique is introduced based on adaptive rather than fixed transforms. Fast parameter adaptation allows training to be part of coding. The proposed technique provides higher peak-signal-to-noise ratio than JPEG with a moderate increase in computational complexity, and neural network techniques.
Keywords :
adaptive signal processing; computational complexity; data compression; image coding; neural nets; JPEG; adaptive cascade architectures; computational complexity; image compression; neural network techniques; parameter adaptation; peak signal-to-noise ratio; training;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20030811
