Ask Your Question
0

how to perform face recognition on images are stored in folder?

asked 2016-05-13 03:37:16 -0600

new_to_linux gravatar image

hello everyone,

i have done with capture,detect and save the current capture image in folder but can anyone tell me how to do recognition for multiple images??

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted
0

answered 2016-05-13 04:08:04 -0600

berak gravatar image

if you have stored your images in a folder per person, like:

+- persons
    +- john
       +- img1.png
       +- img2.png
    +- mary
       +- z42341.png
       +- dsc10.png

you can try similar to this:

int train(Ptr<face::FaceRecognizer> classifier, const String &imgdir, const Size &siz)
{
    string last_n("");
    int label(-1);

    Mat images;
    Mat labels;

    vector<String> vec;
    glob(imgdir,vec,true);
    if ( vec.empty())
        return 0;

    for (size_t i=0; i<vec.size(); i++)
    {
        // extract name from filepath:
        String v = vec[i];
        int r1 = v.find_last_of(SEP);
        String v2 = v.substr(0,r1);
        int r2 = v2.find_last_of(SEP);
        String n = v2.substr(r2+1);
        if (n!=last_n)
        {
            last_n=n;
            label++;
        }
        persons[label] = n;

        // process img & add to trainset:
        Mat img=imread(vec[i],0);
        if (img.empty())
        {
               cerr << "bad image " << vec[i] << endl;
               continue;
        }
        resize(img,img,siz);
        images.push_back(img);
        labels.push_back(label);
    }
    return classifier->train(images, labels);
}

////
Ptr<face::FaceRecognizer> facerec = face::createLBPHFaceRecognizer();
train(facerec, "persons", Size(100,100));

please also have a look at http://docs.opencv.org/ref/master/tut...

edit flag offensive delete link more

Comments

how collecting faces for training ??

new_to_linux gravatar imagenew_to_linux ( 2016-05-15 23:58:24 -0600 )edit

Question Tools

1 follower

Stats

Asked: 2016-05-13 03:37:16 -0600

Seen: 228 times

Last updated: May 23 '16