Ask Your Question
1

Reduce maximal Features on Android.

asked 2013-06-26 07:56:26 -0500

stetro gravatar image

Hi there, is it possible to reduce the feature count of FAST, ORB, STAR or BRISK with the OpenCV4Android Library?

The usual Feature detection looks like this:

private static MatOfKeyPoint detectFeatures(Mat inputPicture) {
    FeatureDetector featureDetector = FeatureDetector.create(FEATURE_DETECTOR_METHOD);
    MatOfKeyPoint matOfKeyPoint = new MatOfKeyPoint();
    featureDetector.detect(inputPicture, matOfKeyPoint);
    return matOfKeyPoint;
}

But is it possible to setup any Settings?

Thanks

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted
2

answered 2013-06-27 11:04:17 -0500

Guanta gravatar image

Yes it is possible, however a little bit awkward: you have to create a parameter.xml (or .yml) file and load it, see e.g. http://answers.opencv.org/question/11478/grid_orb-detector-on-android/ or http://answers.opencv.org/question/12429/matching-orb-descriptors-with-flann-lsh-on-android/, where it is done similarly.

To further reduce your features, you can look at the keypoints-response values and select only e.g. the 500 top ones.

edit flag offensive delete link more
Login/Signup to Answer

Question Tools

Stats

Asked: 2013-06-26 07:56:26 -0500

Seen: 132 times

Last updated: Jun 27 '13