new python filter for STAINED GLASS

[dinasset]

Warning: do not use any gray (R=G=B) colour for "Colour of the lead junctions", otherwise you may get a flat colored image

but this happens randomly, so maybe you will get the correct outcome also with a gray nuance (re-opened gimp and retried, it works....bah!)

[rich2005]

For linux users.

I had a search around but could not find the cartoonizer plugin compiled for linux. It will be there somewhere but to save a hunt, try the attached.
Usual guarantees, there are none

cartoonizer on its own is a nice little plugin, it does work here (32 and 64 bit) and in this machine, Gimp 2.8.18 + Kubuntu 16.04 + gmic 1.7.7pre + v3 Stained Glass

poppies.jpg (327.13 KiB) Viewed 937 times

For the other dependencies
Bevel Reflect is a .scm script, so thats ok
ML Bevel Reflect is available in 32/64 bit linux versions at gimpscripts.com

[dinasset]

those dependencies are for Windows users only (and obsolete, somehow; they were not updated)
the filter detects whether Windows is the OS running or not: if not, it calls another function

Code:

if (systemrunning == "Windows"):
        #  cartonizer with removal of small areas
        pdb.plug_in_cartoonizer(newImage, Preprocessed_Layer,
                # nr-of-colors, colorweight, weightingmode, algorithm
                inNrCol, 8, 2, 0,
                # B&W, legend, denoise
                False, False, True,
                # spotsize, despotting-iterat, erode, erode-iterat
                100, 5, True, 20,       
                # dolines, color, removeshorter, postprocess
                False, 0, 0, 0, 0)
    else:
        #[G'MIC] Cartoon: -cartoon 10,400,30,0.1,1.5,7,0
        pdb.plug_in_gmic(newImage, Preprocessed_Layer, 1,    # 1=input only active layer
        "-v - -cartoon 10,400,30,0.1,1.5,"
        +str(inNrCol)
        +"0")


[dinasset]

Dependencies (updated):
- for all users G'MIC latest version
- for Windows users only: Gimp plug-ins: - plug-in cartoonizer.exe
- for other OS users: - none

[RJKD]

Thanks to this filter I now, at last, begin to understand what constitutes a good image for manipulation.