phistomerging

Perfoms priority region merging based on histogram correlation.

Synopsis

Description

phistomerging merges connected regions of the input region map rg_in if the correlation between their histogram is greater than the specified threshold.

The algorithm uses the priority merging that consists in merging regions with the greater correlation.

The output region map reg_out defines the new regions and the output graph gr_out defines the new relationship between regions.

The correlation between two histograms is given by:

   correlation(H1,H2) = H1.H2 / (norm(H1).norm(H2))

where H1.H2 = scalar product between H1 and H2 and norm(Hi) = euclidean norm of histogram Hi.

The greater is the correlation, the closer is the two histograms.

Parameters

number specifies the number of allowed merging. If number = -1 then all possible merging are done.
threshold specifies the maximum difference allowed between two regions to decide to merge them. Values belongs to [0..1], where 1 corresponds to two equal histograms. Typical values are close to 1 (e.g., 0.7).

Inputs

rg_in: a region map.
gr_in: a graph.
im_in: a grayscale image.

Outputs

rg_out: a region map.
gr_out: a graph.

Result

Returns the number of merging.

Examples

Merges regions yielded by a quadtree splitting process:

   puniformityquadtree 0.9 tangram.pan a.pan
   prg2gr a.pan b.pan
   phistomerging -1 0.94 a.pan b.pan tangram.pan c.pan d.pan

C++ prototype

Errc PHistoMerging(Reg2d &rg_in, Graph2d &gr_in, Img2duc &im_in, Reg2d &rg_out, Graph2d &gr_out, long number, float threshold);

Version française

Fusion prioritaire de régions selon la corrélation d'histogramme.

Author: Laurent Quesnel