Face Detection Isolating faces

Hi guys, I have recently tried the viola jones cascedeobjectdetector to detect faces and it works great. The only thing that I am having trouble is to crop out the faces, isolate and save them individually. have used the icrop function but still stuck thinking of a method to isolate them.. Anyone have any idea?

[Jonathan]

At the moment I can only do one lame thing which is this. I=imread('1.jpg'); %I = rgb2gray(I); faceDetector = vision.CascadeObjectDetector(); bbox_face = step(faceDetector, I);

I_faces = insertObjectAnnotation(I, 'rectangle', bbox_face, 'Face'); figure, imshow(I_faces), title('Detected faces');

for i=1: length(bbox_face) Icrop= imcrop(I,bbox_face(i,:));

Y= Icrop; if i==1 imwrite(Y,'cropped.jpg'); elseif i==2 imwrite(Y,'cropped2.jpg' ); elseif i==3 imwrite(Y,'cropped3.jpg'); else imwrite(Y,'cropped4.jpg' );

end end

Do advice if theres a better way :D

Answer 1

[Anandakumar Selvaraj]

you can try this code to crop the face

faceBBox     = step(faceDetector,videoFrame);

    x = faceBBox(1); y = faceBBox(2); w = faceBBox(3); h = faceBBox(4);
    bboxPolygon = [x, y, x+w, y, x+w, y+1.56*h, x, y+1.56*h];
    face = imcrop(videoFrame,faceBBox);