Ask Your Question

Concept of Cascade Classifier

asked 2016-09-01 12:16:53 -0500

zvone gravatar image

I am trying to undrestand conceptually how does Cascade Classifier training works in OpenCV using LBP. I understand that AdaBoost is used for choosing weak classifiers and combining them to make a strong classifier. I also understand that LBP is used as visual descriotor of features. But the thing I could not find out is how are weak classifiers actually created from LBP? Is there any learning algorithm used, like Support vector machine? Thanks for help.

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2016-09-02 03:23:50 -0500

Well, each LBP feature that was selected during training is used to create binary decision trees. In the default case, these are stumps (1 layer decisions) with a decision weight on the score calculated by the LBP feature. Combining several stumps/decision trees lead to a weak stage.

This is described in full detail in OpenCV 3 Blueprints, Chapter 5!

edit flag offensive delete link more
Login/Signup to Answer

Question Tools

1 follower


Asked: 2016-09-01 12:16:53 -0500

Seen: 135 times

Last updated: Sep 02 '16