Class AlgorithmScaleSaliency

  • All Implemented Interfaces:
    java.awt.event.ActionListener, java.awt.event.WindowListener, java.lang.Runnable, java.util.EventListener

    public class AlgorithmScaleSaliency
    extends AlgorithmBase
    Original source code in C and MATLAB is Copyright 1998-2004 by Timor Kadir Version 1.5 Kadir/Brady Feature detector (Scale Saliency) Code For non-commercial use only Ported by to Java by William Gandler References: 1.) Saliency, Scale, and Image Description Timor Kadir and Michael Brady. International Journal of Computer Vision. 45 (2):83-105, November 2001. 2.) Scale, Saliency, and Scene Description Timor Kadir Department of Engineering Science Robotics Research Group, University of Oxford, Trinity Term, 2002. 3.) An affine invariant salient region detector Timor Kadir, Andrew Zisserman, and Michael Brady, European Conference on Computer Vision 2004. Pages 228 - 241.
    • Field Detail

      • ANTI_ALIASED_SAMPLING_MODE

        private static final int ANTI_ALIASED_SAMPLING_MODE
        See Also:
        Constant Field Values
      • mode

        private int mode
      • startScale

        private int startScale
      • stopScale

        private int stopScale
      • nbins

        private int nbins
      • sigma

        private double sigma
      • wt

        private double wt
      • yt

        private double yt
      • fastplog

        private boolean fastplog
      • plogpArray

        private double[] plogpArray
      • xDim

        private int xDim
      • yDim

        private int yDim
      • sliceSize

        private int sliceSize
    • Constructor Detail

      • AlgorithmScaleSaliency

        public AlgorithmScaleSaliency()
        AlgorithmScaleSaliency - default constructor.
      • AlgorithmScaleSaliency

        public AlgorithmScaleSaliency​(ModelImage srcImg,
                                      int startScale,
                                      int stopScale,
                                      int mode,
                                      int nbins,
                                      double sigma,
                                      double wt,
                                      double yt)
        Parameters:
        srcImg - Source image
        startScale -
        stopScale -
        mode -
        nbins - Number of bins (has 256 for Parzen window PDF estimation)
        sigma - for Parzen window
        wt - Threshold on inter-scale saliency values
        yt - Fraction of saliency maximum value used as threshold