Title :
A 2-D IIR neural hybrid filter for image processing
Author :
Muneyasu, Mitsuji ; Tsujii, Satoshi ; Hinamoto, Takao
Author_Institution :
Fac. of Eng., Hiroshima Univ., Japan
Abstract :
A 2-D IIR neural hybrid (2DINH) filter is developed for image processing. This 2DINH filter consists of the cascade connection of a 2-D IIR linear filter followed by a neural filter. The proposed 2DINH filter can be applied to smoothing for the images corrupted with the Gaussian noise and/or impulsive noise and to edge detection for the images corrupted with the Gaussian noise. The parameters of a neural filter are adjustable by a learning algorithm to adapt itself to the property of an image to be processed. Finally, some examples are given to illustrate the utility of the proposed filter
Keywords :
Gaussian noise; IIR filters; delay circuits; edge detection; image processing; neural nets; smoothing methods; two-dimensional digital filters; 2D IIR neural hybrid filter; Gaussian noise; cascade connection; corrupted images; edge detection; image processing; impulsive noise; learning algorithm; linear filter; smoothing; Finite impulse response filter; Gaussian noise; IIR filters; Image edge detection; Image processing; Information filtering; Neural networks; Noise reduction; Nonlinear filters; Smoothing methods;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on
Conference_Location :
Adelaide, SA
Print_ISBN :
0-7803-1775-0
DOI :
10.1109/ICASSP.1994.389512
