polar grid

Syntax:

      set polar grid {<theta_segments>, <radial_segments>}
                     { qnorm {<power>} | gauss | cauchy | exp | box | hann }
                     { kdensity } { scale <scale> }
                     {theta [min:max]} {r [min:max]}

The polar grid settings are used in conjunction with the plot style with surface to generate a heat map from a set of polar coordinate points. The surface consists of a grid filling a circle divided into segments formed by discrete ranges on theta and r.

Each segment is assigned a value derived from the input set of individual scattered points [x,y,z] by applying a filter operation. The default filter is qnorm 1, which averages each point's z value weighted by the inverse of the point's distance from the center of that grid segment.

Alternative filter operations gauss, cauchy, exp, box, and hann are described in more detail elsewhere. See dgrid3d.

kdensity: This keyword tells the program to use the weighted sum of contributions from all points rather than the weighted average.

scale: This scale facter (default 1.0) is applied to all distances prior to using them in the weighting calculation.

Masking: All input points are used to calculate grid values. The full gridded surface always spans theta range [0:360] and the radial defined by autoscaling or by a previous set rrange command. However the portion of the surface that actually appears in the plot can be restricted to a truncated wedge bounded by lower and upper limits on theta and r. Theta limits must be given in degrees.

For example the following commands will generate a plot which is auto-scaled in size to show all input points. The contributions of all input points are summed, not averaged (kdensity), and only a wedge of the resulting gridded surface is displayed.

     set rrange [0:*]
     set polar grid qnorm kdensity theta [0:190]
     plot DATA with surface, DATA with points