Re: jpeg artifacts, the Glasbey LUT

Posted by Jeremy Adler on Mar 24, 2016; 4:59pm
URL: http://imagej.273.s1.nabble.com/Re-jpeg-artifacts-the-Glasbey-LUT-tp5016000p5016007.html

Curtis and Gabriel,

I have attached a pair of what I have called 'random' LUTs - they are the same except that one has zero set to black.
They are .txt  files as my attempts to send .lut  or .xls were rejected.

They are not completely random and were intended to display incrementally different intensities in contrasting colours - what the glasbey does.

No LUT can guarantee that adjacent pixels with different intensities in an image will have appreciably different display colours, since any two intensities could end up in adjacent pixels and the two most similar colours in an 8 bit LUT will not be that dissimilar on the screen.
In theory, but not using LUTs, we can colour an image with only 4 different colours. But that is difficult computationally. However if we chose 10, I am guessing here, colours it could be done if each ROI had fewer than 9 neighbours: it would require starting with one ROI, colouring it, moving onto adjacent ROIs and for each of these checking their adjacent ROIs and picking a colour etc.

I mentioned using 'random' LUTs in an rather old article  Adler, Microscopy and Analysis Jan 1996, 5-7.

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Curtis Rueden
Sent: den 24 mars 2016 15:57
To: [hidden email]
Subject: Re: jpeg artifacts, the Glasbey LUT

Hi Gabriel,

> I am not sure what the inverted version is. It does not look like
> being in reverse order (which is perhaps what one would expec?) or the
> negative of the original either.  Curtis seems to have added this:
> https://github.com/fiji/fiji/blob/master/luts/glasbey_inverted.lut so
> he should be able to clarify what it does.

For discussion, see:
    https://github.com/fiji/fiji/pull/133

The main goal is to have black (instead of white) be the first color, while preserving the other attributes of Glasbey. However, I actually did not think hard about whether inverting the LUT actually serves that purpose (qualitatively, adjacent colors seem "less different" to my eye).

Regards,
Curtis

--
Curtis Rueden
LOCI software architect - http://loci.wisc.edu/software
ImageJ2 lead, Fiji maintainer - http://imagej.net/User:Rueden Did you know ImageJ has a forum? http://forum.imagej.net/


On Thu, Mar 24, 2016 at 7:11 AM, Gabriel Landini <[hidden email]>
wrote:

> On Thursday 24 Mar 2016 11:15:29 you wrote:
> > Your macro does the job but I am confused by the reference to the
> > glasbey LUT
>
> > My version of FIJI (1.50h) has a glasbey inverted LUT, it this what
> > you meant or is there a glasbey LUT somewhere else ?
>
> Hi Jeremy,
> I once contacted Prof. Glasbey asking about an 8 bit version of his LUT.
> His
> paper did not contain all 256 entries:
> http://onlinelibrary.wiley.com/doi/10.1002/col.20327/pdf
> He kindly sent me the rgb values, so I have been distributing with the
> morphology plugins. I am attaching it here.
>
> I am not sure what the inverted version is. It does not look like
> being in reverse order (which is perhaps what one would expec?) or the
> negative of the original either.  Curtis seems to have added this:
> https://github.com/fiji/fiji/blob/master/luts/glasbey_inverted.lut
> so he should be able to clarify what it does.
>
> Cheers
>
> Gabriel
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html