SVM classifier with sliding window

Helloo!!
I am working on detecting eyes open or close by using SVM classifier with HOG feature and there is problem in running the sliding window and do not know what is the mistake.
as you see the code below , i have already trained SVM classifier.
can anyone help me! please
code:
load classifier
imdsTest = imageDatastore('TestImages');
testImg = readimage(imdsTest,1);
figure,
imshow(testImg)
title('Test Image');
%%
s = 5;
w = 100;
h = 40;
[r,c] = size(testImg);
H = zeros(r,c);
r = floor((r-h)/s)+1;
c = floor((c-w)/s)+1;
%%
r1 = 1;
c1 = 1;
for n=1:r
for m=1:c
img = testImg (c1:c1+39,r1:r1+99);
%imshow(img)
hog1 = extractHOGFeatures(img,'CellSize',[2,2]);
[pred,scores] = predict(classifier,hog1);
if (string(pred) == 'closedEye')
pause(1);
H(c1,r1) = - scores(2);
end
r1 = r1+s;
end
r1 = 1;
c1 = c1 +s;
end
[rw,c1]= find(H>0.75);
figure,
imshow(testImg)
title('closed eye Detection')
for d=1:length(rw)
pasue(1)
imrect(gca, [c1(d), rw(d),100,40]); %
end

1 Comment

Hi, it would be helpful if you could paste the error that you are seeing.

Sign in to comment.

Answers (0)

Asked:

on 4 Mar 2021

Commented:

on 9 Mar 2021

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!