imagesc

I = imagesc(mat; x=, y=, hdr=, proj4=, wkt=, GI=, clim=, cmap=, kw...) --> GMTimage

keywords: GMT, Julia, grid, image

imagesc takes a Float matrix or a GMTgrid type and scales it (by default) to the [0, 255] interval. In the process it creates a GMTimage type. Those types can account for coordinates and projection information, hence the optional arguments. Contrary to its Matlab cousin, it doesn't display the result (that we easily do with imshow(mat)) but return instead a GMTimage object.

• clim: Specify clims as a two-element vector of the form [cmin cmax], where values of the scaled image less than or equal to cmin are assigned that value. The same goes for cmax.

• cmap: If provided, cmap is a GMTcpt and its contents is converted to the GMTimage colormap.

• GI: This can be either a GMTgrid or a GMTimage and its contents is used to set spatial contents (x,y coordinates) and projection info that one may attach to the created image result. This is a handy alterative to the x=, y=, proj4=... options.

• stretch: This option is indicated to select an interval of the range of the z values and use only those to scale to the [0 255] interval. A stretch=true automatically determines good values for histogram stretching via a call to histogram. The form stretch=(zmin,zmax) allows specifying the input limits directly. A previous plot of histogram(mat, show=true) can help determine good values. Note that when this option stretch is used, ALL OTHER options are ignored. See also the rescale function.

If mat is instead a UInt16 GMTimage type we call rescale(I, stretch=true, type=UInt8) instead of issuing an error. In this case clim can be a two elements vector to specify the desired stretch range. The default is to let histogram guess these values.

Examples

Create a grayscale image out of the peaks function.

using GMT

viz(imagesc(GMT.peaks()))

See Also

grdimage, grdview, rescale