gabor filters matlab,matlab实现gabor filter (1)

在 matlab中进行计算gabor filter，自己一步步的构建滤波器矩阵：

先构建gabor滤波器，然后和需要进行滤波的图像进行卷积

代码中的size用来确定gabor filter卷积模板窗口大小。但是这里的size的卷积模板有点偏小，一般都是乘上个系数，或者4或者8.

% GABORFILTER Bi-dimensional Gabor filter with DC component compensation.

% [G,GABOUT]=GABORFILTER(I,S,F,W,P) filters the input image I with the 2D

% Gabor filter G described by the parameters S, F, W and P to create the

% output filtered image GABOUT.

% This version of the 2D Gabor filter is basically a bi-dimensional

% Gaussian function centered at origin (0,0) with variance S modulated by

% a complex sinusoid with polar frequency (F,W) and phase P described by

% the following equation:

%

% G(x,y,S,F,W,P)=k*Gaussian(x,y,S)*(Sinusoid(x,y,F,W,P)-DC(F,S,P)),

% where:

% Gaussian(x,y,S)=exp(-pi*S^2*(x^2+y^2))

% Sinusoid(x,y,F,W,P)=exp(j*(2*pi*F*(x*cos(W)+y*sin(W))+P)))

% DC(F,S,P)=exp(-pi*(F/S)^2+j*P)

%

% PS: The term DC(F,S,P) compensates the inherent DC component produced

% by the Gaussian envelop as shown by Movellan in [1].

%

% Tips:

% 1) To get the real part and the imaginary part of the complex

% filter output use real(gabout) and imag(gabout), respectively;

%

% 2) To get the magnitude and the phase of the complex filter output

% use abs(gabout) and angle(gabout), respectively.

%

% References:

% [1] Movellan, J. R. - Tutorial on Gabor Filters. Tech. rep., 2002.

function [G,GABOUT]=gabor_a(I,S,F,W,P)

if isa(I,'double')~=1

I=double(I);

end

size=fix(1.5/S); % exp(-1.5^2*pi) < 0.1%

%k=2*pi*S^2;

%F=S^2/sqrt(2*pi);

k=1;

for x=-size:size

for y=-size:size

G(size+x+1,size+y+1)=k*exp(-pi*S^2*(x*x+y*y))*...

(exp(j*(2*pi*F*(x*cos(W)+y*sin(W))+P))-exp(-pi*(F/S)^2+j*P));

end

end

GABOUT=conv2(I,double(G),'same');