Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Hello Everyone
I wasn't going to buy into this Photoshop discussion, but I can't
resist!
We change around the colours of our image channels all the time for
various purposes. As long as each of your original images channels are
separated (either as a 24bit RGB or as separate grayscale images) it is
really easy.
First open up a new file of the right size (eg 512 pixels square) and
set it up as an RGB image. Don't worry about channels at all - the are
irrelevant at this stage.
Then select all (CTRL-A) the first image you want to use and copy it
into a new layer on your new image. It should be a grayscale image. Call
the layer some name so you know what it is (it's easy to forget a month
later when you have to adjust something - just double click on the layer
name to change it). Then hit CTRL-U which brings up a dialog with a
whole lot of clour related stuff. Down in the lower right corner is a
little checkbox, called "colourise" - select that and then use the
various sliders to pick your hue, its saturation and lightness and so on
until it looks about what you want. We generally go for 100% saturation,
but reduce the lightness quite a bit so that the highlights are not
saturated. You can make your image any colour you like cyan, yellow,
cerise, maroon ....
Then do the same for each of your other image channels. If you have only
two, you'll should end up with an image with three layers: an empty
background layer and then your two image layers. The are all sorts of
ways of blending them, which can be tricky. But using the "Screen"
option is usally the best. Doubel click on the top layer and choose
screen. Then if you wany to adjust things, you have a couple of sliders
with something like "Blend if..." labels. Choose gray, rather than any
colour channel and move the little triangle markers on the top row
("this layer") - these triangles effectively set a threshold for the
blending operation.
How the final blend ends up depends on two things: (1) the contrast -
brightness of each layer, especially the top one. Adjust this with the
"Adjust levels" dialogue (CTRL-L). Be careful - once you have moveed
these around and saved thechanges, you can't get the orignal back.
Always do these adjustments on a new copy of the image! (there is a
"Duplicate Image" command in the main "Image" menu to do this easily).
.. and (2) something really good that lots of folk miss: if you hold
down the ALT key when you pointer is on one these little triangles, they
separate into two halves, which you can then spread apart - this gives
you really fine control over the blending, by sort of grading the
thresholds.
All of this assumes you really want to blend the images so that the
colours add to each other. IF you want overlays that don't add to each
other, use the "Normal" mode for belnding - this works quite a differnt
way but you use the same controls...
You may need to cycle through a few goes at with the levels and belnding
thrshold until you get it right... then merge all the layers into one
(the command is on the right side of the layers window) and save in
whatever format you want. In the end we save to CMYK for printing, since
this is the colour space that the printers use. To get good prints, you
have to know the relations between the colour balance of your monitor
and the printer - you can use all sorts of fancy profilers to do this -
we just do a few test prints, and write down the settings of what works
best, and then don't change anything!
Hope that helps - all this stuff is in the Photoshop manuals, but it's
bloody hard to work it all out. The real trick with Photoshop is
realising that most of what it does boils down to two things: setting
brightness-contrast ranges for each set of pixels, and then assigning a
color and a transparency to them. Once you have you basic colours for
each image channel, all the rest is pretty much CTRL-L and the layer
blending...
I probably spend as much time showing people how to do all this as I
spend showing them how use the microscopes in the first place. May be we
(ie the list) should put together a little compendium of Photoshop
Bargains for Time-Strapped Digital Microscopists?
IAN
Michael Cammer wrote:
>
> Search the CONFOCAL archive at
> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
> The trick to use in Photoshop to operate in RGB space where most of us
> think better is to type control-y to see the result in CMYK while
> operating in RGB.
>
> Practically, to do CFP/YFP merge in RGB space would be to put the CFP in
> the blue channel, the YFP in the red channel, and the maximum pixel
> combination of the two channels in the green channel.
>
> ___________________________________
> WORK: http://www.aecom.yu.edu/aif/
> PERSONAL: http://www.art-studio.us/
>
> On Wed, 26 Feb 2003, Glen MacDonald wrote:
>
> > Search the CONFOCAL archive at
> > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> >
> > Starting fresh from the original grayscales is a fairly painless means
> > of generating CMYK figures. The problems usually arise from generating
> > an RGB figure then converting it to CMYK. The color conversion can be
> > dreadful and affected by your color management module (if any), OS or
> > application version and how far you've manipulated colors in the RGB
> > colorspace. The one manner in which I've found this to work in
> > Photoshop 7 is to select a CMYK printer profile for the working space,
> > and 'Preserve embedded profiles' for color management policy. At least
> > this works in Photoshop 7, with the Tektronix Phaser SDX Colorsync
> > profile and using the Colorsync conversion engine in Apple's OSX.
> > Adobe changes the location and names of these settings between software
> > versions. Havent' yet compared other CMYK printer profiles or other
> > color conversion engines.
> >
> > Regards,
> > Glen
> > On Wednesday, February 26, 2003, at 03:15 PM, Jacqui Ross wrote:
> >
> > > Search the CONFOCAL archive at
> > > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> > >
> > > Hi Hugo,
> > >
> > > I'm not sure how you were working with the files. If you have already
> > > tried
> > > using the channels menu with ghastly results, read no further. If
> > > not, it's
> > > worth giving it a go, although it's a bit hard when I don't know what
> > > your
> > > images look like.
> > >
> > > Make a new file with the same resolution as the images with a black
> > > background
> > > and in CMYK colour mode. Make your original images grayscale and
> > > adjust if
> > > needed using the levels menu. Then go to the new black image and go to
> > > the
> > > channels menu. For the cyan (CFP) image, turn off all the channels
> > > except cyan
> > > and make sure this channel is highlighted. Go back to your cyan (CFP)
> > > image,
> > > select all and copy. Then go to the black image and paste the image
> > > into the
> > > cyan channel.
> > >
> > > For the YFP one, do the same thing except this time, turn all the
> > > channels off
> > > except yellow and paste the YFP image into the yellow channel. Then
> > > turn all
> > > channels on again and you will have your composite in CMYK.
> > >
> > > Cheers,
> > >
> > > Jacqui.
> > >
> > > "Caldas, Hugo" wrote:
> > >
> > >> Search the CONFOCAL archive at
> > >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> > >>
> > >> I have a question that mind sound a little stupid and not really what
> > >> this
> > >> mailing list is about but....
> > >>
> > >> I have acquired CFP and YFP images, and am trying to use Photoshop to
> > >> overlay the 2. Does anyone know how I can do this keeping the colors
> > >> cyan
> > >> and yellow? I have attempted CMYK but didn't take me very far. My
> > >> composite
> > >> images just end up looking horrible and I cannot make my point this
> > >> way.
> > >> Any help would be greatly appreciated.
> > >>
> > >> Hugo Caldas
> > >> Columbus Children's Hospital
> > >> Columbus, OH
> > >>
> > >> -----Original Message-----
> > >> From: Guy Cox [mailto:[log in to unmask]]
> > >> Sent: Wednesday, February 26, 2003 3:44 AM
> > >> To: [log in to unmask]
> > >> Subject: YFP/CFP excitation
> > >>
> > >> Search the CONFOCAL archive at
> > >> http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
> > >>
> > >> This reference recently posted to the MPLSM group is absolutely
> > >> essential reading for those engaged in the current debate about
> > >> MP excitation of GFPs.
> > >>
> > >> Essentially (and as expected since its a non-symmetrical chromophore)
> > >> the excitation spectrum for the various GFP versions is a close
> > >> match to the single photon spectrum (at twice the wavelength).
> > >> Excitation of eGFP at 800nm is only about 20% of the excitation
> > >> at the peak, 860nm. Unfortunately they don't give wtGFP but
> > >> since that has a major excitation peak at 395 is could be good
> > >> at 790-800 in MP for those who have non-tunable lasers.
> > >>
> > >> Thanks to Ben Youn for posting this.
> > >>
> > >> Guy
> > >>
> > >>> Here's a reference for TPE spectra for a couple of fluorophores,
> > >>> including eGFP, eCFP, eYFP and DsRed
> > >>>
> > >>> Two-photon excitation action cross-sections of the autofluorescent
> > >>> proteins
> > >>> Blab GA, Lommerse PHM, Cognet L, Harms GS, Schmidt T
> > >>> CHEMICAL PHYSICS LETTERS 350 (1-2): 71-77 DEC 14 2001
> > >>>
> > >>> Regards,
> > >>> Ben Youn
> > >>
> > >> Assoc. Prof. Guy Cox, ooOOOOOOoo
> > >> E.M. Unit, F09 # oOOOO | | OOOOo #
> > >> University of Sydney ### OOO| | | | | |OOO ###
> > >> NSW 2006, Australia ### OOO | | | | | | OOO ###
> > >> Ph: 02 9351 3176 ### OO | | | | | | | | OO ###
> > >> Fax: 02 9351 7682 ##### | | | | | | | | #####
> > >> ==#####============================#####==
> > >> http://get.to/gcc ##### #####
> > >> http://guycox.cjb.net ~~#####~~~~~~~~~~~~~~~~~~~~~~~~~~~~#####~~
> > >
> > > Jacqueline Ross
> > > Biomedical Imaging Research Unit
> > > Division of Anatomy with Radiology
> > > Faculty of Medical & Health Sciences
> > > The University of Auckland
> > > Private Bag 92019
> > > Auckland, NEW ZEALAND
> > >
> > > Tel: 64 9 373 7599 Ext 87438
> > > Fax: 64 9 373 7484
> > >
> > > http://www.health.auckland.ac.nz/biru/
> > >
> > Glen MacDonald
> > Core for Communication Research
> > Virginia Merrill Bloedel Hearing Research Center
> > Box 357923
> > University of Washington
> > Seattle, WA 98195-7923 USA
> > (206) 616-4156
> > [log in to unmask]
> >
> > ************************************************************************
> > ******
> > The box said "Requires Windows 95 or better", so I bought a Macintosh.
> > ************************************************************************
> > ******
> >
--
Professor Ian Gibbins
Anatomy & Histology
Flinders University of South Australia
GPO Box 2100, Adelaide, SA 5001
Australia
Phone: +61-8-8204 5271
FAX: +61-8-8277 0085
Email: [log in to unmask]
|
