Title :
Image compression using multi-layer neural networks
Author :
Abdel-Wahhab, Osama ; Fahmy, Moustafa M.
Author_Institution :
Dept. of Electr. Eng., King Fahd Univ. of Pet. & Miner., Dhahran, Saudi Arabia
Abstract :
A new neural network data compression method is presented. This work extends the use of 2-layer neural networks to multi-layer networks. Results show the performance superiority of multi-layer neural networks compared to that of the 2-layer one especially at high compression ratios. To overcome the long training time required for multi-layered networks, a recently developed training algorithm has been used. A modified feedback error is proposed to further reduce the training time and to enhance the image quality. Also, a redistribution of the gray levels in the training phase is proposed to make the minimization of the mean square error more related to the human visual system
Keywords :
data compression; feedforward neural nets; image coding; learning (artificial intelligence); multilayer perceptrons; transform coding; auto-associative transform coding; data compression; feedforward neural nets; gray levels redistribution; image compression; image quality; mean square error minimisation; modified feedback error; multilayer neural networks; multilayer perceptrons; training algorithm; Compressors; Data compression; Image coding; Image reconstruction; Minerals; Multi-layer neural network; Neural networks; Petroleum; Transform coding; Vectors;
Conference_Titel :
Computers and Communications, 1997. Proceedings., Second IEEE Symposium on
Conference_Location :
Alexandria
Print_ISBN :
0-8186-7852-6
DOI :
10.1109/ISCC.1997.615992