Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.

Image Feature Detection with LBP Operator?

Fred12Fred12 Member Posts: 344 Unicorn
edited November 2018 in Help

edit: Had to edit the subject title, something got messed up...

 

I want to do image feature detection (and object detection) based on LBP feature recognition, so I basically want to binarize images based on some threshold, and then do LBP to recognize so called "visual words", which are then put together in a characteristic histogram that can be processed / classified by a SVM for object detection.

Is there any tutorial on this, or some example from where to start from?

Tagged:

Best Answer

  • VaclavVaclav RapidMiner Certified Expert, Member Posts: 23 Maven
    Solution Accepted

    Hello Fred,

    there is Fast LBP Detector operator with tutorial process. You can start there.

     

    Best wishes,

    Vaclav

Answers

  • Fred12Fred12 Member Posts: 344 Unicorn

    I have a serious problem... none of the tutorial processes is really working...

    I used positive and neg. examples for eyes and haar detector with random forest inside, but no eyes were detected:

     

    Unbenannt4.PNG

     

    but I basically had problems with every other of the tutorial p rocesses, too. None of them were basically working like it should..

  • VaclavVaclav RapidMiner Certified Expert, Member Posts: 23 Maven

    Hello Fred,

    the example serves for illustration how to setup a process. If you need accurate detection, you need better training set. Also LBP is not best detector.

     

    Best wishes,

    Vaclav

Sign In or Register to comment.