Title :
Autoassociative neural networks for image compression: a massively parallel implementation
Author_Institution :
Lab. de Traitement des Signaux, Ecole Polytech. Federale de Lausanne, Switzerland
fDate :
31 Aug-2 Sep 1992
Abstract :
A massively parallel implementation on the associative string processor (ASP) machine of a neural-network-based technique for image compression is presented. Despite the linear structure of the ASP and the use of fixed arithmetic for the implementation, promising results are shown in terms of learning speed, on the order of 109 connections per second, and the quality of the reconstructed images
Keywords :
data compression; image processing; neural chips; parallel architectures; associative string processor machine; autoassociative neural networks; fixed arithmetic; image compression; learning speed; linear structure; reconstructed image quality; Application specific processors; Arithmetic; Data communication; Electronic mail; Image coding; Image reconstruction; Neural networks; Neurons; Vectors; Video compression;
Conference_Titel :
Neural Networks for Signal Processing [1992] II., Proceedings of the 1992 IEEE-SP Workshop
Conference_Location :
Helsingoer
Print_ISBN :
0-7803-0557-4
DOI :
10.1109/NNSP.1992.253675
