How to improve the result of the performance of the gabor filter

feats=gabfilter(fn1,25,[0.05 0.1],[6 6]);

Answer 1

[Yusuf Suer Erdem]

Example below is a gabor filter example. When you change the wavelength and the orientation. You ll get images with different qualities.

.tif file

clc; clear; close all;

I = imread('board.tif');
I = rgb2gray(I);

wavelength = 4;
orientation = 90;
[mag,phase] = imgaborfilt(I,wavelength,orientation);

tiledlayout(1,3)
nexttile
imshow(I)
title('Original Image')
nexttile
imshow(mag,[])
title('Gabor Magnitude')
nexttile

[ASMA MOHAMMAD]

I used a gabor filter on a total of photos, and the recognition rate on the photos was 64, and I want to improve the luxury ratio

close all

clear all

imgpath{4}='C:\Users\tosh\Documents\pro1\;

fd4=fopen('FFF.txt','rt'); % 294 subjects Train Total train = 493

fet=[];

fet2=[];

for i=1:165

txt1=fscanf(fd4,'%s\n',1);

for j=1:6

fn1=sprintf('%s/%s middle/0%iROI.jpg',imgpath{4},txt1,j) %u01000s0001_fnf1.jpg

img1=imread(fn1);

feats=gabfilter(fn1,27,[0.05 0.5],[7 7]);

glr(:,:)=feats;

fet=reshape(glr,size(glr,1)*size(glr,2),1);

fet2 = [fet2 fet;];

end

[Yusuf Suer Erdem]

I really could't comprehend all the codes but it must be realted to parameters which are in gabfilter() function.

[ASMA MOHAMMAD]

function Output = gabfilter(image,N,freq,partition)

stage = partition(1);

orientation = partition(2);

im = im2double(imread(image));

[W, H] = size(im);

for j = 1:orientation

Gr = Gabor(N,[0 j],freq,partition,1);

switch j

case 1

ima1 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

case 2

ima2 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

case 3

ima3 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

case 4

ima4 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

case 5

ima5 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

case 6

ima6 = ifft2(fft2(im).*fft2(im2double(Gr),size(im,1),size(im,2)));

end

end

ima = zeros(W, H);

for p = 1:W

for q = 1:H

ima(p, q) = (minimum(ima1(p, q), ima2(p, q), ima3(p, q), ima4(p, q), ima5(p, q), ima6(p, q))) * 256/6;

end

end

Output = ima;

end

function index = minimum(a, b, c, d, e, f)

res = min(a, b);

res = min(res, c);

res = min(res, d);

res = min(res, e);

res = min(res, f);

switch res

case a

index = 1;

case b

index = 2;

case c

index = 3;

case d

index = 4;

case e

index = 5;

case f

index = 6;

end

end

[yanqi liu]

sir，may be upload Gabor.m for some analysis