Title :
A high performance adaptive image compression system using a generative neural network: DynAmic Neural Network II (DANN II)
Author :
Rios, Andres ; Kabuka, Mansur R.
Author_Institution :
Dept. of Electr. & Comput. Eng., Miami Univ., Coral Gables, FL, USA
Abstract :
The system is guaranteed theoretically to compress to any feasible rate, with as low a distortion rate as required. It also exhibits user selectable compression and error rates, ability to compress general data types, and adaptation to the data source. The compression system is based on a novel family of connectionist algorithms and generative algorithms used in conjunction with new artificial neural network models that permit the determination of a quasi-optimal architecture for compressing a given data source
Keywords :
adaptive systems; data compression; image processing; neural nets; DANN II; DynAmic Neural Network II; adaptive image compression system; artificial neural network models; connectionist algorithms; generative algorithms; generative neural network; quasi-optimal architecture; Adaptive systems; Artificial neural networks; Computer architecture; Data compression; Error analysis; Image coding; Neural networks; Neurons; Programmable control; Rate distortion theory;
Conference_Titel :
Data Compression Conference, 1993. DCC '93.
Conference_Location :
Snowbird, UT
Print_ISBN :
0-8186-3392-1
DOI :
10.1109/DCC.1993.253129
