Title :
Application of multi-layer neural networks to image compression
Author :
Ahmed, Osama A. ; Fahmy, M.M.
Author_Institution :
King Fahd Univ. of Pet. & Miner., Saudi Arabia
Abstract :
A new neural network data compression is presented. This work extends the use of 2-layer neural networks to multi-layer networks. A special method has been proposed to reduce the training time of such networks. Results show the performance superiority of the proposed multi-layer neural networks compared with that of the 2-layer ones, especially at high compression ratios. A modified feed-back error is proposed to reduce the training time and enhance the image quality
Keywords :
coding errors; data compression; feedforward neural nets; image coding; image resolution; learning (artificial intelligence); multilayer perceptrons; transform coding; autoassociative transform coding; error scaling; feedforward neural network; high compression ratio; image compression; image quality; modified feed-back error; multi-layer neural networks; multilayer perceptron; performance superiority; training time reduction; Compressors; Decoding; Feedforward systems; Image coding; Image converters; Multi-layer neural network; Multilayer perceptrons; Neural networks; Neurons; Training data;
Conference_Titel :
Circuits and Systems, 1997. ISCAS '97., Proceedings of 1997 IEEE International Symposium on
Print_ISBN :
0-7803-3583-X
DOI :
10.1109/ISCAS.1997.622069
