Title :
Neural network technology for image compression
Author_Institution :
Bolton Inst., UK
Abstract :
The paper presents an extensive survey on the development of neural network technology for image compression. Three main categories are covered. These include neural networks directly developed for image compression, neural network implementation of traditional algorithms and development of neural network based technology which provides further improvements over the existing image compression algorithms. The author concludes that neural networks still have a long way to go before substantial improvement over traditional image compression algorithms can be achieved
Keywords :
adaptive codes; backpropagation; fractals; image coding; neural nets; vector quantisation; wavelet transforms; image compression; implementation; neural network technology; traditional algorithms;
Conference_Titel :
Broadcasting Convention, 1995. IBC 95., International
Conference_Location :
Amsterdam
Print_ISBN :
0-85296-644-X
DOI :
10.1049/cp:19950961