Digital Signal Processing Reference
In-Depth Information
Fig. 8.2
Amplitude responses of the filters obtained by the various competitive algorithms
Ta b l e 8 . 1
Parameter settings
for the IIWO algorithm
Parameters
Value
Initial Population
40
Maximum population
40
Maximum no. of seeds
10
Minimum no. of seeds
0
Std. dev. for seed dispersal
0.1-0.001
Reward parameter,
α
0.25
Penalty parameter,
γ
0.8
2D filters [
9
,
21
,
22
]. The original image is first corrupted with Gaussian noise of
mean zero and variance 0
.
005. Then the image is transformed into the frequency
domain using Fast Fourier Transform (FFT) and multiplied with the filter transfer
function. The filtered image is then obtained by taking the inverse FFT (IFFT). The
image processing techniques are performed using the MATLAB image processing
toolbox. The results of image denoising by the filters designed by the competitive
algorithms are shown in Fig.
8.3
.
Search WWH ::
Custom Search