python lbp image classification

Question

I am working on a personal project: gender classification (male | female) in python. I'm beginner in this domain

I computed histograms for every image in training data. Now, to test if a test image is male or female is possible to make an average histogram for male | female and compare test histograms? Or I must compare all histograms with test histogram?

If there is possible to make an average. How should I do it? Also, is ok to use SVM for classification?

PS. I am looking for free faces databases.

Thanks

Mathias · Accepted Answer · 2017-02-23 14:59:01Z

1

If you want to use SVM as a classifier it does not make a lot of sense to make one average histogram for male and one for female because when you train you SVM classifier you can take all the histograms into account, but if you compute the average histograms you can use a nearest neighbor classifier instead.

answered Feb 23, 2017 at 14:59

Mathias

1631 gold badge1 silver badge8 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Razvan Over a year ago

Thanks for your answer.

Razvan Over a year ago

I am wondering if i computed the average right. Considering a histogram as an array, the average histogram is the sum of each image array divided by the number of images?

Mathias Over a year ago

I think it sounds right, you have computed separate histograms for each image, then summing each bin in the histogram separately and then dividing by the number of images used in the summing. That is how you should compute the mean histogram

Collectives™ on Stack Overflow

python lbp image classification

1 Answer 1

3 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Related