image processing
Locked
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal I have a simple question, related to an annoying problem. I have an image of a multilabelled fluorescent specimen, displayed as a merged RGB image. I would like to analyze just the red channel, so I split the channels and obtain B & W images of the individual component. They all look as bright as the original. However, if I only have a single red image and would like to convert it to a B & W image (change mode RGB to grey), for 8-bit analysis in ImageJ, for example, the resulting image is very dull and fainter. Furthermore, this effect is much more dramatic with a red image than with a green image. What is happening? Thanks, Judy Judy Trogadis Bio-Imaging Coordinator St. Michael's Hospital, 7Queen 30 Bond St. Toronto, ON M5B 1W8, Canada ph: 416-864-6060 x6337 pager: 416-685-9219 fax: 416-864-6043 [hidden email] |
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Dear Judy: They use a formula to convert RGB to gray intensities. I am not sure what this formula is, but it takes into account weighted contributions from each channel. So if you have pure red (255) and nothing else in an RGB image and convert it into an 8-bit gray you get an intensity not 255 but only 85 (which tells you, by the way, the Red comes with coefficient 0.3). Michael Model, Ph.D. Confocal Microscopy Core Dpt. Biological Sciences Kent State University Kent, OH 44242 tel. 330-672-2874 -----Original Message----- From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Judy Trogadis Sent: Monday, April 07, 2008 11:03 AM To: [hidden email] Subject: image processing Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal I have a simple question, related to an annoying problem. I have an image of a multilabelled fluorescent specimen, displayed as a merged RGB image. I would like to analyze just the red channel, so I split the channels and obtain B & W images of the individual component. They all look as bright as the original. However, if I only have a single red image and would like to convert it to a B & W image (change mode RGB to grey), for 8-bit analysis in ImageJ, for example, the resulting image is very dull and fainter. Furthermore, this effect is much more dramatic with a red image than with a green image. What is happening? Thanks, Judy Judy Trogadis Bio-Imaging Coordinator St. Michael's Hospital, 7Queen 30 Bond St. Toronto, ON M5B 1W8, Canada ph: 416-864-6060 x6337 pager: 416-685-9219 fax: 416-864-6043 [hidden email] |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Not 0.3 but 1/3 of course -----Original Message----- From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Judy Trogadis Sent: Monday, April 07, 2008 11:03 AM To: [hidden email] Subject: image processing Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal I have a simple question, related to an annoying problem. I have an image of a multilabelled fluorescent specimen, displayed as a merged RGB image. I would like to analyze just the red channel, so I split the channels and obtain B & W images of the individual component. They all look as bright as the original. However, if I only have a single red image and would like to convert it to a B & W image (change mode RGB to grey), for 8-bit analysis in ImageJ, for example, the resulting image is very dull and fainter. Furthermore, this effect is much more dramatic with a red image than with a green image. What is happening? Thanks, Judy Judy Trogadis Bio-Imaging Coordinator St. Michael's Hospital, 7Queen 30 Bond St. Toronto, ON M5B 1W8, Canada ph: 416-864-6060 x6337 pager: 416-685-9219 fax: 416-864-6043 [hidden email] |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hye all, There is a lot if good information about that subject in the microscopy site. http://www.microscopyu.com/articles/digitalimaging/colorbalance.html John Russ, is also a man with lot of practical experience in this field. He wrote a book on Image processing and several good reviews on color photography. Bye, Patrick Van Oostveldt Quoting Judy Trogadis <[hidden email]>: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have a simple question, related to an annoying problem. > > I have an image of a multilabelled fluorescent specimen, displayed > as a merged RGB image. I would like to analyze just the red channel, > so I split the channels and obtain B & W images of the individual > component. They all look as bright as the original. > > However, if I only have a single red image and would like to convert > it to a B & W image (change mode RGB to grey), for 8-bit analysis > in ImageJ, for example, the resulting image is very dull and > fainter. Furthermore, this effect is much more dramatic with a red > image than with a green image. > > What is happening? > Thanks, > Judy > > > Judy Trogadis > Bio-Imaging Coordinator > St. Michael's Hospital, 7Queen > 30 Bond St. > Toronto, ON M5B 1W8, Canada > ph: 416-864-6060 x6337 > pager: 416-685-9219 > fax: 416-864-6043 > [hidden email] > -- Dep. Moleculaire Biotechnologie Coupure links 653 B 9000 GENT tel 09 264 5969 fax 09 264 6219 |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hi Judy, Make sure you treat your red image just the same as any merged image: start with an RGB image (not indexed color) and work with a single channel rather than converting the image mode from RGB to grey. Good luck, Gudrun On Apr 7, 2008, at 11:02 AM, Judy Trogadis wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have a simple question, related to an annoying problem. > > I have an image of a multilabelled fluorescent specimen, displayed as > a merged RGB image. I would like to analyze just the red channel, so I > split the channels and obtain B & W images of the individual > component. They all look as bright as the original. > > However, if I only have a single red image and would like to convert > it to a B & W image (change mode RGB to grey), for 8-bit analysis in > ImageJ, for example, the resulting image is very dull and fainter. > Furthermore, this effect is much more dramatic with a red image than > with a green image. > > What is happening? > Thanks, > Judy > > > Judy Trogadis > Bio-Imaging Coordinator > St. Michael's Hospital, 7Queen > 30 Bond St. > Toronto, ON M5B 1W8, Canada > ph: 416-864-6060 x6337 > pager: 416-685-9219 > fax: 416-864-6043 > [hidden email] > > ____________________________________________ Gudrun Ihrke, Ph.D. Research Assistant Professor Department of Pharmacology (C2025) Uniformed Services University School of Medicine 4301 Jones Bridge Road Bethesda, MD 20814 phone office: (301) 295 3225 phone lab: (301) 295 3221 FAX: (301) 295 3220 email: [hidden email] |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Things to investigate in your software:
Your microscope should be able to write individual channels as black and white (pure intensity) images, probably as TIFF files. You should not have to convert an RGB image to its component parts. Can you look at a single channel in black and white? Eyeballs evaluate intensities better in black and white than in color. How does your software handle more than three channel images? Unless you need to display more than one color on the same image, don't present fluorescent images in monochromatic color. -- Paul Herzmark Specialist [hidden email] Department of Molecular and Cell Biology 479 Life Science Addition University of California, Berkeley Berkeley, CA 94720-3200 (510) 643-9603 (510) 643-9500 fax On Mon, Apr 7, 2008 at 8:02 AM, Judy Trogadis <[hidden email]> wrote: I have a simple question, related to an annoying problem. |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
-
Judy, With Image J, if you convert an RGB image to monochrome, the gray value you obtain for any given pixel is just a simple average of the red, green, and blue values. If your starting image is an RGB image with just red, you should split the channels with image J, and keep the red channel. It should be exactly the same as the original. The green and blue channels should be black (zero value). If you just convert the RGB to 8-bit, you divide the intensity by three (since you average your red value with the values for green and blue, namely zero and zero). -- Julio Vazquez Fred Hutchinson Cancer Research Center Seattle, WA 98109-1024 On Apr 7, 2008, at 8:02 AM, Judy Trogadis wrote:
|
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Thank you to all who replied - I knew it was simple. Simply converting RGB to 8-bit does not work. It divides the signal and if you have images in all 3 channels, it's total confusion! I feel embarrassed, I should have known this - I admit it. Cheers, Judy >>> Julio Vazquez <[hidden email]> 04/07/08 1:16 PM >>> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal - Judy, With Image J, if you convert an RGB image to monochrome, the gray value you obtain for any given pixel is just a simple average of the red, green, and blue values. If your starting image is an RGB image with just red, you should split the channels with image J, and keep the red channel. It should be exactly the same as the original. The green and blue channels should be black (zero value). If you just convert the RGB to 8-bit, you divide the intensity by three (since you average your red value with the values for green and blue, namely zero and zero). -- Julio Vazquez Fred Hutchinson Cancer Research Center Seattle, WA 98109-1024 On Apr 7, 2008, at 8:02 AM, Judy Trogadis wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have a simple question, related to an annoying problem. > > I have an image of a multilabelled fluorescent specimen, displayed > as a merged RGB image. I would like to analyze just the red > channel, so I split the channels and obtain B & W images of the > individual component. They all look as bright as the original. > > However, if I only have a single red image and would like to > convert it to a B & W image (change mode RGB to grey), for 8-bit > analysis in ImageJ, for example, the resulting image is very dull > and fainter. Furthermore, this effect is much more dramatic with a > red image than with a green image. > > What is happening? > Thanks, > Judy > > > Judy Trogadis > Bio-Imaging Coordinator > St. Michael's Hospital, 7Queen > 30 Bond St. > Toronto, ON M5B 1W8, Canada > ph: 416-864-6060 x6337 > pager: 416-685-9219 > fax: 416-864-6043 > [hidden email] |
|
|
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Don't feel embarrassed. Your question and the replies got me thinking and realizing that I needed to bone up on image processing myself. I went to Amazon.com and ordered the 5th edition of John Russ's Handbook.
So thanks for pushing me to review the basics.
Mollie Lange
International Center for Spinal Cord Injury
Kennedy Krieger Institute
Baltimore, Maryland >>> Judy Trogadis <[hidden email]> 4/7/2008 2:17 PM >>> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Thank you to all who replied - I knew it was simple. Simply converting RGB to 8-bit does not work. It divides the signal and if you have images in all 3 channels, it's total confusion! I feel embarrassed, I should have known this - I admit it. Cheers, Judy >>> Julio Vazquez <[hidden email]> 04/07/08 1:16 PM >>> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal - Judy, With Image J, if you convert an RGB image to monochrome, the gray value you obtain for any given pixel is just a simple average of the red, green, and blue values. If your starting image is an RGB image with just red, you should split the channels with image J, and keep the red channel. It should be exactly the same as the original. The green and blue channels should be black (zero value). If you just convert the RGB to 8-bit, you divide the intensity by three (since you average your red value with the values for green and blue, namely zero and zero). -- Julio Vazquez Fred Hutchinson Cancer Research Center Seattle, WA 98109-1024 On Apr 7, 2008, at 8:02 AM, Judy Trogadis wrote: > Search the CONFOCAL archive at > http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have a simple question, related to an annoying problem. > > I have an image of a multilabelled fluorescent specimen, displayed > as a merged RGB image. I would like to analyze just the red > channel, so I split the channels and obtain B & W images of the > individual component. They all look as bright as the original. > > However, if I only have a single red image and would like to > convert it to a B & W image (change mode RGB to grey), for 8-bit > analysis in ImageJ, for example, the resulting image is very dull > and fainter. Furthermore, this effect is much more dramatic with a > red image than with a green image. > > What is happening? > Thanks, > Judy > > > Judy Trogadis > Bio-Imaging Coordinator > St. Michael's Hospital, 7Queen > 30 Bond St. > Toronto, ON M5B 1W8, Canada > ph: 416-864-6060 x6337 > pager: 416-685-9219 > fax: 416-864-6043 > [hidden email] Disclaimer: The materials in this e-mail are private and may contain Protected Health Information. Please note that e-mail is not necessarily confidential or secure. Your use of e-mail constitutes your acknowledgment of these confidentiality and security limitations. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying, distribution, or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this e-mail in error, please immediately notify the sender via telephone or return e-mail. |
|
|
In reply to this post by Judy Trogadis
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
As others have pointed out, this happens because in a simple conversion
of colour to grey-scale red has a low weighting. But you shouldn't be
doing that ....
If your red image is stored as an rgb data set (24 bit) just separate the
channels exactly as you did with your rgb image. You well get two blank
images for green and blue and the third will be what you want.
If your image is stored as paletted colour (8-bit, 256 colours) you can select
'convert to rgb' and then split the channels as above. Or you can just replace
the palette with a linear grey palette. Make sure you select 'maintain indexes' or
equivalent, NOT 'use closest value'. If your imaging programme doesn't have a
linear grey palette in its list, save the palette from a good-quality grey-scale
image to get one.
Guy
Optical Imaging Techniques in Cell Biology
by Guy Cox CRC Press / Taylor & Francis http://www.guycox.com/optical.htm ______________________________________________ Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________ Phone +61 2 9351 3176 Fax +61 2 9351 7682 Mobile 0413 281 861 ______________________________________________ http://www.guycox.net From: Confocal Microscopy List on behalf of Judy Trogadis Sent: Tue 08/04/08 1:02 AM To: [hidden email] Subject: image processing Search the CONFOCAL archive at |
|
|
To point out what is probably obvious, I believe that this procedure will only work if working with a maximum of three labels, red, green and blue. Only then can you separate the merged image into the RGB component channels. However, Judy wrote that she has multilabeled specimens - perhaps more than three. Thus, any additional fluorophore would be labeled with a LUT that would be some combination of Red, Green and Blue. Separating the Red channel of a such an RGB merged image would also separate out the red component of a second or third fluorophore. Just my 2 cents worth.
George
George Ring, Ph.D.
Dept. of Cell and Developmental Biology SUNY Upstate Medical University 750 E. Adams St. Syracuse NY 13210 Tel. (315) 464-8595 FAX (315) 464-8535 email: [hidden email] >>> Guy Cox <[hidden email]> 4/7/08 8:22:15 PM >>> Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
As others have pointed out, this happens because in a simple conversion
of colour to grey-scale red has a low weighting. But you shouldn't be
doing that ....
If your red image is stored as an rgb data set (24 bit) just separate the
channels exactly as you did with your rgb image. You well get two blank
images for green and blue and the third will be what you want.
If your image is stored as paletted colour (8-bit, 256 colours) you can select
'convert to rgb' and then split the channels as above. Or you can just replace
the palette with a linear grey palette. Make sure you select 'maintain indexes' or
equivalent, NOT 'use closest value'. If your imaging programme doesn't have a
linear grey palette in its list, save the palette from a good-quality grey-scale
image to get one.
Guy
Optical Imaging Techniques in Cell Biology
by Guy Cox CRC Press / Taylor & Francis http://www.guycox.com/optical.htm ______________________________________________ Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________ Phone +61 2 9351 3176 Fax +61 2 9351 7682 Mobile 0413 281 861 ______________________________________________ http://www.guycox.net From: Confocal Microscopy List on behalf of Judy Trogadis Sent: Tue 08/04/08 1:02 AM To: [hidden email] Subject: image processing Search the CONFOCAL archive at |
|
|
In reply to this post by Guy Cox
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Dear List,
Does the recent version of Photoshop allow re-mapping to [0,4095],
[0,65535], based on zero, linear (Log) stretch, etc.?
Is there a free plug-in to read 12-bit images in Photoshop? I heard that
John Russ's handbook had a supllement CD for extra price with lots of different
but useful plug-ins, any feedback would be appreciated?
Cheers,
Vitaly
NCI-Frederick,
301-846-6575
|
|
|
In reply to this post by George Ring
Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
Please remove me from the list. Thankyou Thomas Pitta Ph.D. Micro Video Instruments Digital Imaging Specialist 508 580 0080 ex 241 508 330 8356 cell From: Confocal
Microscopy List [mailto:[hidden email]] On Behalf Of George Ring To point out what
is probably obvious, I believe that this procedure will only work if working
with a maximum of three labels, red, green and blue. Only then can
you separate the merged image into the RGB component channels.
However, Judy wrote that she has multilabeled specimens - perhaps more than
three. Thus, any additional fluorophore would be labeled with a
LUT that would be some combination of Red, Green and Blue. Separating the
Red channel of a such an RGB merged image would also separate out the red
component of a second or third fluorophore. Just my 2 cents worth. George George Ring,
Ph.D. Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal As others have pointed out, this happens because in a simple
conversion of colour to grey-scale red has a low weighting. But you
shouldn't be doing that .... If your red image is stored as an rgb data set (24 bit) just
separate the channels exactly as you did with your rgb image. You
well get two blank images for green and blue and the third will be what you want. If your image is stored as paletted colour (8-bit, 256 colours)
you can select 'convert to rgb' and then split the channels as above. Or
you can just replace the palette with a linear grey palette. Make sure you select
'maintain indexes' or equivalent, NOT 'use closest value'. If your imaging
programme doesn't have a linear grey palette in its list, save the palette from a good-quality
grey-scale image to get one.
Guy Optical Imaging Techniques in Cell Biology http://www.guycox.com/optical.htm From: Confocal Microscopy List on behalf of Judy Trogadis Search the CONFOCAL archive at |
| Free forum by Nabble | Edit this page |