Title :
Edge-preserving neural network based image restoration
Author :
Bao, Paul ; Wang, Dianhui
Author_Institution :
Dept. of Comput., Hong Kong Polytech., Kowloon, Hong Kong
Abstract :
This paper presents a combined approach for image restoration with edge-preserving regularization, subband coding, and artificial neural network. The multilayer perceptron model is employed to implement the restoration of images. The main merit of the neural network model is its massive parallelism with strong robustness for transmission noise and parameter or structure perturbation. The experiment has shown that the proposed approach outperforms SPIHT on both objective and subjective quality.
Keywords :
image coding; image restoration; multilayer perceptrons; transform coding; wavelet transforms; SPIHT; artificial neural network; edge-preserving neural network; edge-preserving regularization; experiment; image restoration; massive parallelism; multilayer perceptron model; neural network model; objective quality; parameter perturbation; structure perturbation; subband coding; subjective quality; transmission noise robustness; wavelet transform; Artificial neural networks; Computer networks; Degradation; Discrete wavelet transforms; Image coding; Image reconstruction; Image restoration; Neural networks; Nonlinear equations; Wavelet transforms;
Conference_Titel :
Signals, Systems, and Computers, 1999. Conference Record of the Thirty-Third Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-7803-5700-0
DOI :
10.1109/ACSSC.1999.831856
