Neural network approaches to fractal image compression and decompression

Citation
Kt. Sun et al., Neural network approaches to fractal image compression and decompression, NEUROCOMPUT, 41, 2001, pp. 91-107
Citations number
33
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
NEUROCOMPUTING
ISSN journal
09252312 → ACNP
Volume
41
Year of publication
2001
Pages
91 - 107
Database
ISI
SICI code
0925-2312(200110)41:<91:NNATFI>2.0.ZU;2-X
Abstract
In image compression technologies, fractal image compression/decompression has the advantages of a high compression ratio and a low loss ratio. Howeve r, it requires a great deal of computation, which limits its applications, and so far, no parallel processing technique has been designed and implemen ted. In this study, we use neural networks to perform a large number of com putations in fractal image compression and decompression in parallel. The s imulation results show that the quality of images generated by neural netwo rks is similar to that produced using traditional methods, which verifies t he high value of our research, which has shown that the neural network tech nology is useful and efficient when applied to fractal image compression an d decompression. (C) 2001 Elsevier Science B.V. All rights reserved.