GREG2 : Two Dimensional Problems

We call ``two dimensional data'' any data represented by a function f(X,Y). In practice, this representation may be regularly or randomly sampled. The first case corresponds to the Regular Grid array for which RG=f(I,J) and linear conversion formulae exist to convert between I and X and J and Y respectively. The second corresponds to triplets (X,Y,Z) stored in the corresponding buffers. Basic handling of 2-D data includes contouring, perspective view, bitmap processing, and extraction of one dimensional data.

Only regularly sampled 2-D data (i.e. a Regular Grid) can be conveniently handled. The limits of a Regular Grid can be defined as the plot limits using command LIMITS /RGDATA. In this case, GREG computes the user coordinates corresponding to pixels 0.5 and NX+0.5 in X (same in Y) and defines these values as the user coordinate limits. This means that the Box will exactly cover the Image corresponding to the given map.

For randomly sampled data, GREG provides a way of interpolating them on a Regular Grid by command RANDOM_MAP. All other commands are provided to work exclusively on the Regular Grid array. This includes :

• A general contouring facility using command RGMAP
• A simple perspective view using command PERSPECTIVE
• A bitmap display capability for X-Window terminals and PostScript output, command PLOT
• A resampling facility either to produce finer contours, or to produce maps with the same sampling, and hence comparable pixel by pixel.
• Limited map analysis facilities : pixel value interrogation with command DRAW VALUE, statistics using command MEAN, masking part of a map using command MASK, computations of global and local extrema using command EXTREMA, strip extraction using command STRIP. For more general image processing facilities, the user is referred to the GILDAS documentation.

All these commands belong to the ``GREG2'' language.