Class SparseScoreRectifiedNcc<T extends ImageGray<T>>


public class SparseScoreRectifiedNcc<T extends ImageGray<T>> extends DisparitySparseRectifiedScoreBM<float[],T>
Compute NCC error for sparse disparity. Should produce similar results to dense version. Image normalization is computed using local statistics instead of global statistics across input image.
  • Field Details

    • statsLeft

      public final boofcv.alg.disparity.block.SparseScoreRectifiedNcc.SparseStatistics statsLeft
    • statsRight

      public final boofcv.alg.disparity.block.SparseScoreRectifiedNcc.SparseStatistics statsRight
    • eps

      public float eps
    • normalizeInput

      public boolean normalizeInput
    • scoreLtoR

      protected float[] scoreLtoR
    • scoreRtoL

      protected float[] scoreRtoL
  • Constructor Details

    • SparseScoreRectifiedNcc

      public SparseScoreRectifiedNcc(int blockRadiusX, int blockRadiusY, Class<T> imageType)
  • Method Details

    • configure

      public void configure(int disparityMin, int disparityRange)
      Description copied from class: DisparitySparseRectifiedScoreBM
      Configures the disparity search
      configure in class DisparitySparseRectifiedScoreBM<float[],T extends ImageGray<T>>
      disparityMin - Minimum disparity that it will check. Must be ≥ 0 and < disparityMax
      disparityRange - Number of possible disparity values estimated. The max possible disparity is min+range-1.
    • scoreDisparity

      protected void scoreDisparity(int disparityRange, boolean leftToRight)
      Description copied from class: DisparitySparseRectifiedScoreBM
      Scores the disparity using image patches.
      Specified by:
      scoreDisparity in class DisparitySparseRectifiedScoreBM<float[],T extends ImageGray<T>>
      disparityRange - The local range for disparity
      leftToRight - If true then the disparity is being from in left to right direction (the typical)