ImageJ

Saturation and Quantifying

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

14 messages Options

NatyC

Saturation and Quantifying

Hello everyone!

I'm working on a project that has me quantifying a specific color in the image.
I was wondering if ImageJ had a way of detecting colors that aren’t
visible to our eye, but are a part of the image.
(Since I'm working with a stain, it’s not present unless its under a
fluorescent light.)
Also, does anyone have recommendations on how to quantify colors in
using ImageJ?

Thanks,
Natalia

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

Herbie-2

Re: Saturation and Quantifying

Natalia,

you write:
"[...] detecting colors that aren’t visible to our eye, but are a part
of the image."

Interesting, but what do you really mean by this?

If you have a digital image, what I assume, because you want to do
digital image processing with ImageJ, then you have pixels that in fact
are large matrices of numbers that can be visualized on a computer
monitor as shades of gray or as shades of red, green, and blue
respectively. You may even use false color to visualize these numbers.

If however, what you call images are in fact objects that are to be
imaged, like by the human eye or a photographic device, then things
become much clearer. There are cameras that can catch infrared or
ultraviolet light outside the spectral sensitivity of the human eye.

So what's the situation?

Best

Herbie

_________________________________________
On 25.10.12 20:26, Natalia Chacon wrote:

> Hello everyone!
>
> I'm working on a project that has me quantifying a specific color in the image.
> I was wondering if ImageJ had a way of detecting colors that aren’t
> visible to our eye, but are a part of the image.
> (Since I'm working with a stain, it’s not present unless its under a
> fluorescent light.)
> Also, does anyone have recommendations on how to quantify colors in
> using ImageJ?
>
> Thanks,
> Natalia
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

Vivek Saraswat-2

Re: Saturation and Quantifying

In reply to this post by NatyC

Hi Natalia,

From what I understand from your question is that:

You have some sort of a image having different colours blue green red
orange of different shades and you want to find out that area fraction does
each colour occupy?

If I am right, do write back, I can help you with an algorithm.

Regards
*Vivek Saraswat*
*Pre-final year Undergraduate*
*Department of Materials Science and Engineering*
*IIT Bombay (IIT B), India*
☎ *+91-9969924136*
[hidden email]
http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>

On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon <[hidden email]
> wrote:

> Hello everyone!
>
> I'm working on a project that has me quantifying a specific color in the
> image.
> I was wondering if ImageJ had a way of detecting colors that aren’t
> visible to our eye, but are a part of the image.
> (Since I'm working with a stain, it’s not present unless its under a
> fluorescent light.)
> Also, does anyone have recommendations on how to quantify colors in
> using ImageJ?
>
> Thanks,
> Natalia
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

NatyC

Re: Saturation and Quantifying

Hi Vivek,

Yes, I'm trying to find the area and the amount that a certain colour
occupies in my whole image and then in each section. (I'm viewing an
eye with that is stained to locate a protein.)

If there is an algorithm that can assist me with this process I'm up
for trying it!

Thanks,
Natalia

On 10/25/12, Vivek Saraswat <[hidden email]> wrote:

> Hi Natalia,
>
> From what I understand from your question is that:
>
> You have some sort of a image having different colours blue green red
> orange of different shades and you want to find out that area fraction does
> each colour occupy?
>
> If I am right, do write back, I can help you with an algorithm.
>
> Regards
> *Vivek Saraswat*
> *Pre-final year Undergraduate*
> *Department of Materials Science and Engineering*
> *IIT Bombay (IIT B), India*
> ☎ *+91-9969924136*
> [hidden email]
> http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>
>
>
>
> On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
> <[hidden email]
>> wrote:
>
>> Hello everyone!
>>
>> I'm working on a project that has me quantifying a specific color in the
>> image.
>> I was wondering if ImageJ had a way of detecting colors that aren’t
>> visible to our eye, but are a part of the image.
>> (Since I'm working with a stain, it’s not present unless its under a
>> fluorescent light.)
>> Also, does anyone have recommendations on how to quantify colors in
>> using ImageJ?
>>
>> Thanks,
>> Natalia
>>
>> --
>> 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

Vivek Saraswat-2

Re: Saturation and Quantifying

Hi Natalia,

So here's the solution:

1. In Image J, open your image and then run 8-bit colour mode. It would ask
how many colours you want to digitize the image into? Here you would have
to use your approximation. Like, say 8 colours.

2. The RGB values into which the image has been digitized can be viewed
from the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB
values

3. Image-->Type-->RGB Colour. Make as many duplicates of this image as no.
of colours into which image is digitized.

4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue box,
set Thresholding method as “Default”, Thresholding Colour as “B&W” (Black
and White) and Colour Space as “RGB”. Set Colour thresholds on RGB values
equal to their corresponding values from LUT table.

5. Now make each of the images binary and then analyze to get the data
about no. of particles of each colour, size, feret etc.

I know this may sound a bit confusing so you can write back if anyrhing is
not clear. To summarize, our aim here is to simplify the quantification
process so we first make a guess about no. of colours in which we want to
digitize it. Then in 8-bit color mode, we can specify how many colours? And
then we find up their RGB values from LUT Table. We then convert the image
back to RGB so that we can again threshold it. Before thresholding we make
no. of copies of this RGB image=No. of colours. Then for each image, we
threshold it with corresponding value from LUT Table, then binarize each of
them and analyze them.

If you want to repeat this process a number of times, you have to write a
macro code!.

All the best!

Regards
*Vivek Saraswat*
*Pre-final year Undergraduate*
*Department of Materials Science and Engineering*
*IIT Bombay (IIT B), India*
☎ *+91-9969924136*
[hidden email]
http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>

On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
<[hidden email]>wrote:

> Hi Vivek,
>
> Yes, I'm trying to find the area and the amount that a certain colour
> occupies in my whole image and then in each section. (I'm viewing an
> eye with that is stained to locate a protein.)
>
> If there is an algorithm that can assist me with this process I'm up
> for trying it!
>
> Thanks,
> Natalia
>
> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
> > Hi Natalia,
> >
> > From what I understand from your question is that:
> >
> > You have some sort of a image having different colours blue green red
> > orange of different shades and you want to find out that area fraction
> does
> > each colour occupy?
> >
> > If I am right, do write back, I can help you with an algorithm.
> >
> > Regards
> > *Vivek Saraswat*
> > *Pre-final year Undergraduate*
> > *Department of Materials Science and Engineering*
> > *IIT Bombay (IIT B), India*
> > ☎ *+91-9969924136*
> > [hidden email]
> > http://in.linkedin.com/in/viveksaraswat26<
> http://home.iitb.ac.in/%7Evivek_saraswat>
> >
> >
> >
> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
> > <[hidden email]
> >> wrote:
> >
> >> Hello everyone!
> >>
> >> I'm working on a project that has me quantifying a specific color in the
> >> image.
> >> I was wondering if ImageJ had a way of detecting colors that aren’t
> >> visible to our eye, but are a part of the image.
> >> (Since I'm working with a stain, it’s not present unless its under a
> >> fluorescent light.)
> >> Also, does anyone have recommendations on how to quantify colors in
> >> using ImageJ?
> >>
> >> Thanks,
> >> Natalia
> >>
> >> --
> >> 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
>

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

Vergara, Leoncio A.

Re: Saturation and Quantifying

I think there is an important component to the question that has not been answered and is not clear,
On the original message she wrote: "I was wondering if ImageJ had a way of detecting colors that aren’t visible to our eye, but are a part of the image. (Since I'm working with a stain, it’s not present unless its under a fluorescent light.)"

What is not clear is how the images are captured, if this is a fluorescent stain, then the images should not be handled as if they were simple color images. I guess there would be filter sets involved and then each fluorescent channel would be an independent grayscale image, much easier to quantify... in terms of area and "amount" (then is not clear what she means by "amount", intensity? Optical density?

Leoncio

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Vivek Saraswat
Sent: Thursday, October 25, 2012 4:09 PM
To: [hidden email]
Subject: Re: Saturation and Quantifying

Hi Natalia,

So here's the solution:

1. In Image J, open your image and then run 8-bit colour mode. It would ask how many colours you want to digitize the image into? Here you would have to use your approximation. Like, say 8 colours.

2. The RGB values into which the image has been digitized can be viewed from the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB values

3. Image-->Type-->RGB Colour. Make as many duplicates of this image as no.
of colours into which image is digitized.

4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue box, set Thresholding method as “Default”, Thresholding Colour as “B&W” (Black and White) and Colour Space as “RGB”. Set Colour thresholds on RGB values equal to their corresponding values from LUT table.

5. Now make each of the images binary and then analyze to get the data about no. of particles of each colour, size, feret etc.

I know this may sound a bit confusing so you can write back if anyrhing is not clear. To summarize, our aim here is to simplify the quantification process so we first make a guess about no. of colours in which we want to digitize it. Then in 8-bit color mode, we can specify how many colours? And then we find up their RGB values from LUT Table. We then convert the image back to RGB so that we can again threshold it. Before thresholding we make no. of copies of this RGB image=No. of colours. Then for each image, we threshold it with corresponding value from LUT Table, then binarize each of them and analyze them.

If you want to repeat this process a number of times, you have to write a macro code!.

All the best!

Regards
*Vivek Saraswat*
*Pre-final year Undergraduate*
*Department of Materials Science and Engineering* *IIT Bombay (IIT B), India* ☎ *+91-9969924136* [hidden email] http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>

On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
<[hidden email]>wrote:

> Hi Vivek,
>
> Yes, I'm trying to find the area and the amount that a certain colour
> occupies in my whole image and then in each section. (I'm viewing an
> eye with that is stained to locate a protein.)
>
> If there is an algorithm that can assist me with this process I'm up
> for trying it!
>
> Thanks,
> Natalia
>
> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
> > Hi Natalia,
> >
> > From what I understand from your question is that:
> >
> > You have some sort of a image having different colours blue green
> > red orange of different shades and you want to find out that area
> > fraction
> does
> > each colour occupy?
> >
> > If I am right, do write back, I can help you with an algorithm.
> >
> > Regards
> > *Vivek Saraswat*
> > *Pre-final year Undergraduate*
> > *Department of Materials Science and Engineering* *IIT Bombay (IIT
> > B), India* ☎ *+91-9969924136* [hidden email]
> > http://in.linkedin.com/in/viveksaraswat26<
> http://home.iitb.ac.in/%7Evivek_saraswat>
> >
> >
> >
> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
> > <[hidden email]
> >> wrote:
> >
> >> Hello everyone!
> >>
> >> I'm working on a project that has me quantifying a specific color
> >> in the image.
> >> I was wondering if ImageJ had a way of detecting colors that aren’t
> >> visible to our eye, but are a part of the image.
> >> (Since I'm working with a stain, it’s not present unless its under
> >> a fluorescent light.) Also, does anyone have recommendations on how
> >> to quantify colors in using ImageJ?
> >>
> >> Thanks,
> >> Natalia
> >>
> >> --
> >> 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
>

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

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

Christian Goosmann

Re: Saturation and Quantifying

In reply to this post by NatyC

Natalia Chacon wrote:

Dear Natalia,
if you are speaking of digital images in RGB format, you might want to
explore 'Image>Adjust>Color Threshold...' from the ImageJ menu. There
you can filter out those parts of your image that have a certain color,
saturation, brightness. After segmenting your image with this tool, you
can analyze the areas in terms of number, size etc.
Good luck
Christian

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

NatyC

Re: Saturation and Quantifying

In reply to this post by Vergara, Leoncio A.

Dear Leoncio

The images I'm working with were provided from my mentor, who stated
that the sample had been stained to located a specific protein.
From what I understood the images given to me were taken under a
confocal microscope.

My goal is to quantify the amount of this protein by working off of
"how much of the my image is stained “.

Whenever the image is edited to draw out the color, the ending product
makes it difficult to comprehend how much of the protein is in my
sample.
(To draw out the color I'm doing Image>Adjust>Color Balance. This is
making the staining visible to my eye.)
I'm relying on ImageJ to quantify the amount of color there is for the stain.
This will give me an idea of how much of the protein is in my image

So I'm not looking for the optical density of this image but rather
the amount of specific color.

Thanks,
Natalia

On 10/25/12, Vergara, Leoncio A. <[hidden email]> wrote:

> I think there is an important component to the question that has not been
> answered and is not clear,
> On the original message she wrote: "I was wondering if ImageJ had a way of
> detecting colors that aren’t visible to our eye, but are a part of the
> image. (Since I'm working with a stain, it’s not present unless its under a
> fluorescent light.)"
>
> What is not clear is how the images are captured, if this is a fluorescent
> stain, then the images should not be handled as if they were simple color
> images. I guess there would be filter sets involved and then each
> fluorescent channel would be an independent grayscale image, much easier to
> quantify... in terms of area and "amount" (then is not clear what she means
> by "amount", intensity? Optical density?
>
> Leoncio
>
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Vivek
> Saraswat
> Sent: Thursday, October 25, 2012 4:09 PM
> To: [hidden email]
> Subject: Re: Saturation and Quantifying
>
> Hi Natalia,
>
> So here's the solution:
>
> 1. In Image J, open your image and then run 8-bit colour mode. It would ask
> how many colours you want to digitize the image into? Here you would have to
> use your approximation. Like, say 8 colours.
>
> 2. The RGB values into which the image has been digitized can be viewed from
> the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB values
>
> 3. Image-->Type-->RGB Colour. Make as many duplicates of this image as no.
> of colours into which image is digitized.
>
> 4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue box,
> set Thresholding method as “Default”, Thresholding Colour as “B&W” (Black
> and White) and Colour Space as “RGB”. Set Colour thresholds on RGB values
> equal to their corresponding values from LUT table.
>
> 5. Now make each of the images binary and then analyze to get the data about
> no. of particles of each colour, size, feret etc.
>
> I know this may sound a bit confusing so you can write back if anyrhing is
> not clear. To summarize, our aim here is to simplify the quantification
> process so we first make a guess about no. of colours in which we want to
> digitize it. Then in 8-bit color mode, we can specify how many colours? And
> then we find up their RGB values from LUT Table. We then convert the image
> back to RGB so that we can again threshold it. Before thresholding we make
> no. of copies of this RGB image=No. of colours. Then for each image, we
> threshold it with corresponding value from LUT Table, then binarize each of
> them and analyze them.
>
> If you want to repeat this process a number of times, you have to write a
> macro code!.
>
> All the best!
>
> Regards
> *Vivek Saraswat*
> *Pre-final year Undergraduate*
> *Department of Materials Science and Engineering* *IIT Bombay (IIT B),
> India* ☎ *+91-9969924136* [hidden email]
> http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>
>
>
>
> On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
> <[hidden email]>wrote:
>
>> Hi Vivek,
>>
>> Yes, I'm trying to find the area and the amount that a certain colour
>> occupies in my whole image and then in each section. (I'm viewing an
>> eye with that is stained to locate a protein.)
>>
>> If there is an algorithm that can assist me with this process I'm up
>> for trying it!
>>
>> Thanks,
>> Natalia
>>
>> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
>> > Hi Natalia,
>> >
>> > From what I understand from your question is that:
>> >
>> > You have some sort of a image having different colours blue green
>> > red orange of different shades and you want to find out that area
>> > fraction
>> does
>> > each colour occupy?
>> >
>> > If I am right, do write back, I can help you with an algorithm.
>> >
>> > Regards
>> > *Vivek Saraswat*
>> > *Pre-final year Undergraduate*
>> > *Department of Materials Science and Engineering* *IIT Bombay (IIT
>> > B), India* ☎ *+91-9969924136* [hidden email]
>> > http://in.linkedin.com/in/viveksaraswat26<
>> http://home.iitb.ac.in/%7Evivek_saraswat>
>> >
>> >
>> >
>> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
>> > <[hidden email]
>> >> wrote:
>> >
>> >> Hello everyone!
>> >>
>> >> I'm working on a project that has me quantifying a specific color
>> >> in the image.
>> >> I was wondering if ImageJ had a way of detecting colors that aren’t
>> >> visible to our eye, but are a part of the image.
>> >> (Since I'm working with a stain, it’s not present unless its under
>> >> a fluorescent light.) Also, does anyone have recommendations on how
>> >> to quantify colors in using ImageJ?
>> >>
>> >> Thanks,
>> >> Natalia
>> >>
>> >> --
>> >> 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
>>
>
> --
> 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

Vivek Saraswat-2

Re: Saturation and Quantifying

Hi Natalia,

Didn't the method i prescribed work in your case?

Regards
*Vivek Saraswat*
*Pre-final year Undergraduate*
*Department of Materials Science and Engineering*
*IIT Bombay (IIT B), India*
☎ *+91-9969924136*
[hidden email]
[hidden email]
http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>

On Fri, Oct 26, 2012 at 11:50 PM, Natalia Chacon <[hidden email]
> wrote:

> Dear Leoncio
>
> The images I'm working with were provided from my mentor, who stated
> that the sample had been stained to located a specific protein.
> From what I understood the images given to me were taken under a
> confocal microscope.
>
> My goal is to quantify the amount of this protein by working off of
> "how much of the my image is stained “.
>
> Whenever the image is edited to draw out the color, the ending product
> makes it difficult to comprehend how much of the protein is in my
> sample.
> (To draw out the color I'm doing Image>Adjust>Color Balance. This is
> making the staining visible to my eye.)
> I'm relying on ImageJ to quantify the amount of color there is for the
> stain.
> This will give me an idea of how much of the protein is in my image
>
> So I'm not looking for the optical density of this image but rather
> the amount of specific color.
>
> Thanks,
> Natalia
>
> On 10/25/12, Vergara, Leoncio A. <[hidden email]> wrote:
> > I think there is an important component to the question that has not been
> > answered and is not clear,
> > On the original message she wrote: "I was wondering if ImageJ had a way
> of
> > detecting colors that aren’t visible to our eye, but are a part of the
> > image. (Since I'm working with a stain, it’s not present unless its
> under a
> > fluorescent light.)"
> >
> > What is not clear is how the images are captured, if this is a
> fluorescent
> > stain, then the images should not be handled as if they were simple color
> > images. I guess there would be filter sets involved and then each
> > fluorescent channel would be an independent grayscale image, much easier
> to
> > quantify... in terms of area and "amount" (then is not clear what she
> means
> > by "amount", intensity? Optical density?
> >
> > Leoncio
> >
> >
> > -----Original Message-----
> > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Vivek
> > Saraswat
> > Sent: Thursday, October 25, 2012 4:09 PM
> > To: [hidden email]
> > Subject: Re: Saturation and Quantifying
> >
> > Hi Natalia,
> >
> > So here's the solution:
> >
> > 1. In Image J, open your image and then run 8-bit colour mode. It would
> ask
> > how many colours you want to digitize the image into? Here you would
> have to
> > use your approximation. Like, say 8 colours.
> >
> > 2. The RGB values into which the image has been digitized can be viewed
> from
> > the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB values
> >
> > 3. Image-->Type-->RGB Colour. Make as many duplicates of this image as
> no.
> > of colours into which image is digitized.
> >
> > 4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue
> box,
> > set Thresholding method as “Default”, Thresholding Colour as “B&W” (Black
> > and White) and Colour Space as “RGB”. Set Colour thresholds on RGB values
> > equal to their corresponding values from LUT table.
> >
> > 5. Now make each of the images binary and then analyze to get the data
> about
> > no. of particles of each colour, size, feret etc.
> >
> > I know this may sound a bit confusing so you can write back if anyrhing
> is
> > not clear. To summarize, our aim here is to simplify the quantification
> > process so we first make a guess about no. of colours in which we want to
> > digitize it. Then in 8-bit color mode, we can specify how many colours?
> And
> > then we find up their RGB values from LUT Table. We then convert the
> image
> > back to RGB so that we can again threshold it. Before thresholding we
> make
> > no. of copies of this RGB image=No. of colours. Then for each image, we
> > threshold it with corresponding value from LUT Table, then binarize each
> of
> > them and analyze them.
> >
> > If you want to repeat this process a number of times, you have to write a
> > macro code!.
> >
> > All the best!
> >
> > Regards
> > *Vivek Saraswat*
> > *Pre-final year Undergraduate*
> > *Department of Materials Science and Engineering* *IIT Bombay (IIT B),
> > India* ☎ *+91-9969924136* [hidden email]
> > http://in.linkedin.com/in/viveksaraswat26<
> http://home.iitb.ac.in/%7Evivek_saraswat>
> >
> >
> >
> > On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
> > <[hidden email]>wrote:
> >
> >> Hi Vivek,
> >>
> >> Yes, I'm trying to find the area and the amount that a certain colour
> >> occupies in my whole image and then in each section. (I'm viewing an
> >> eye with that is stained to locate a protein.)
> >>
> >> If there is an algorithm that can assist me with this process I'm up
> >> for trying it!
> >>
> >> Thanks,
> >> Natalia
> >>
> >> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
> >> > Hi Natalia,
> >> >
> >> > From what I understand from your question is that:
> >> >
> >> > You have some sort of a image having different colours blue green
> >> > red orange of different shades and you want to find out that area
> >> > fraction
> >> does
> >> > each colour occupy?
> >> >
> >> > If I am right, do write back, I can help you with an algorithm.
> >> >
> >> > Regards
> >> > *Vivek Saraswat*
> >> > *Pre-final year Undergraduate*
> >> > *Department of Materials Science and Engineering* *IIT Bombay (IIT
> >> > B), India* ☎ *+91-9969924136* [hidden email]
> >> > http://in.linkedin.com/in/viveksaraswat26<
> >> http://home.iitb.ac.in/%7Evivek_saraswat>
> >> >
> >> >
> >> >
> >> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
> >> > <[hidden email]
> >> >> wrote:
> >> >
> >> >> Hello everyone!
> >> >>
> >> >> I'm working on a project that has me quantifying a specific color
> >> >> in the image.
> >> >> I was wondering if ImageJ had a way of detecting colors that aren’t
> >> >> visible to our eye, but are a part of the image.
> >> >> (Since I'm working with a stain, it’s not present unless its under
> >> >> a fluorescent light.) Also, does anyone have recommendations on how
> >> >> to quantify colors in using ImageJ?
> >> >>
> >> >> Thanks,
> >> >> Natalia
> >> >>
> >> >> --
> >> >> 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
> >>
> >
> > --
> > 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
>

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

NatyC

Re: Saturation and Quantifying

Hi Vivek,

I just got to work so I haven't had time to try it out!

Natalia

On 10/26/12, Vivek Saraswat <[hidden email]> wrote:

> Hi Natalia,
>
> Didn't the method i prescribed work in your case?
>
> Regards
> *Vivek Saraswat*
> *Pre-final year Undergraduate*
> *Department of Materials Science and Engineering*
> *IIT Bombay (IIT B), India*
> ☎ *+91-9969924136*
> [hidden email]
> [hidden email]
> http://in.linkedin.com/in/viveksaraswat26<http://home.iitb.ac.in/%7Evivek_saraswat>
>
>
>
> On Fri, Oct 26, 2012 at 11:50 PM, Natalia Chacon
> <[hidden email]
>> wrote:
>
>> Dear Leoncio
>>
>> The images I'm working with were provided from my mentor, who stated
>> that the sample had been stained to located a specific protein.
>> From what I understood the images given to me were taken under a
>> confocal microscope.
>>
>> My goal is to quantify the amount of this protein by working off of
>> "how much of the my image is stained “.
>>
>> Whenever the image is edited to draw out the color, the ending product
>> makes it difficult to comprehend how much of the protein is in my
>> sample.
>> (To draw out the color I'm doing Image>Adjust>Color Balance. This is
>> making the staining visible to my eye.)
>> I'm relying on ImageJ to quantify the amount of color there is for the
>> stain.
>> This will give me an idea of how much of the protein is in my image
>>
>> So I'm not looking for the optical density of this image but rather
>> the amount of specific color.
>>
>> Thanks,
>> Natalia
>>
>> On 10/25/12, Vergara, Leoncio A. <[hidden email]> wrote:
>> > I think there is an important component to the question that has not
>> > been
>> > answered and is not clear,
>> > On the original message she wrote: "I was wondering if ImageJ had a way
>> of
>> > detecting colors that aren’t visible to our eye, but are a part of the
>> > image. (Since I'm working with a stain, it’s not present unless its
>> under a
>> > fluorescent light.)"
>> >
>> > What is not clear is how the images are captured, if this is a
>> fluorescent
>> > stain, then the images should not be handled as if they were simple
>> > color
>> > images. I guess there would be filter sets involved and then each
>> > fluorescent channel would be an independent grayscale image, much
>> > easier
>> to
>> > quantify... in terms of area and "amount" (then is not clear what she
>> means
>> > by "amount", intensity? Optical density?
>> >
>> > Leoncio
>> >
>> >
>> > -----Original Message-----
>> > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
>> Vivek
>> > Saraswat
>> > Sent: Thursday, October 25, 2012 4:09 PM
>> > To: [hidden email]
>> > Subject: Re: Saturation and Quantifying
>> >
>> > Hi Natalia,
>> >
>> > So here's the solution:
>> >
>> > 1. In Image J, open your image and then run 8-bit colour mode. It would
>> ask
>> > how many colours you want to digitize the image into? Here you would
>> have to
>> > use your approximation. Like, say 8 colours.
>> >
>> > 2. The RGB values into which the image has been digitized can be viewed
>> from
>> > the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB
>> > values
>> >
>> > 3. Image-->Type-->RGB Colour. Make as many duplicates of this image as
>> no.
>> > of colours into which image is digitized.
>> >
>> > 4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue
>> box,
>> > set Thresholding method as “Default”, Thresholding Colour as “B&W”
>> > (Black
>> > and White) and Colour Space as “RGB”. Set Colour thresholds on RGB
>> > values
>> > equal to their corresponding values from LUT table.
>> >
>> > 5. Now make each of the images binary and then analyze to get the data
>> about
>> > no. of particles of each colour, size, feret etc.
>> >
>> > I know this may sound a bit confusing so you can write back if anyrhing
>> is
>> > not clear. To summarize, our aim here is to simplify the quantification
>> > process so we first make a guess about no. of colours in which we want
>> > to
>> > digitize it. Then in 8-bit color mode, we can specify how many colours?
>> And
>> > then we find up their RGB values from LUT Table. We then convert the
>> image
>> > back to RGB so that we can again threshold it. Before thresholding we
>> make
>> > no. of copies of this RGB image=No. of colours. Then for each image, we
>> > threshold it with corresponding value from LUT Table, then binarize
>> > each
>> of
>> > them and analyze them.
>> >
>> > If you want to repeat this process a number of times, you have to write
>> > a
>> > macro code!.
>> >
>> > All the best!
>> >
>> > Regards
>> > *Vivek Saraswat*
>> > *Pre-final year Undergraduate*
>> > *Department of Materials Science and Engineering* *IIT Bombay (IIT B),
>> > India* ☎ *+91-9969924136* [hidden email]
>> > http://in.linkedin.com/in/viveksaraswat26<
>> http://home.iitb.ac.in/%7Evivek_saraswat>
>> >
>> >
>> >
>> > On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
>> > <[hidden email]>wrote:
>> >
>> >> Hi Vivek,
>> >>
>> >> Yes, I'm trying to find the area and the amount that a certain colour
>> >> occupies in my whole image and then in each section. (I'm viewing an
>> >> eye with that is stained to locate a protein.)
>> >>
>> >> If there is an algorithm that can assist me with this process I'm up
>> >> for trying it!
>> >>
>> >> Thanks,
>> >> Natalia
>> >>
>> >> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
>> >> > Hi Natalia,
>> >> >
>> >> > From what I understand from your question is that:
>> >> >
>> >> > You have some sort of a image having different colours blue green
>> >> > red orange of different shades and you want to find out that area
>> >> > fraction
>> >> does
>> >> > each colour occupy?
>> >> >
>> >> > If I am right, do write back, I can help you with an algorithm.
>> >> >
>> >> > Regards
>> >> > *Vivek Saraswat*
>> >> > *Pre-final year Undergraduate*
>> >> > *Department of Materials Science and Engineering* *IIT Bombay (IIT
>> >> > B), India* ☎ *+91-9969924136* [hidden email]
>> >> > http://in.linkedin.com/in/viveksaraswat26<
>> >> http://home.iitb.ac.in/%7Evivek_saraswat>
>> >> >
>> >> >
>> >> >
>> >> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
>> >> > <[hidden email]
>> >> >> wrote:
>> >> >
>> >> >> Hello everyone!
>> >> >>
>> >> >> I'm working on a project that has me quantifying a specific color
>> >> >> in the image.
>> >> >> I was wondering if ImageJ had a way of detecting colors that aren’t
>> >> >> visible to our eye, but are a part of the image.
>> >> >> (Since I'm working with a stain, it’s not present unless its under
>> >> >> a fluorescent light.) Also, does anyone have recommendations on how
>> >> >> to quantify colors in using ImageJ?
>> >> >>
>> >> >> Thanks,
>> >> >> Natalia
>> >> >>
>> >> >> --
>> >> >> 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
>> >>
>> >
>> > --
>> > 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
>>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

Stoyan Pavlov

Re: Saturation and Quantifying

In reply to this post by NatyC

Dear Natalia,
As these are confocal images you are talking about, it is wrong to meddle
with the color balance! What you need is to separate the channels
(Image=>Color=>Split channels) and then chose the channel that has your
protein. (If on other hand these aren't multiphoton images (only one marker
is recorded) you may skip this step and just convert the image to either
16-bit or 8-bit grey (If your color images are 48bit RGB, then chose 16bit
grey).
If you are unable to see well the recorded protein you may correct
brightness and contrast, but be careful not to oversaturate the image. In
other words your upper brightness limit (the white point) must be set above
your brightest recorded levels or you will lose information.
Next step is to segment the Image (i.e. Threshold it) so you get a binary
image of "protein" and "background".
Next you can easy calculate occupied surface fraction. (Analyze=>Set
measurements=> check Area and Limit to Threshold boxes, and then run
Measure). When you are familiar with all the steps you can easily record
your own macro (Plugins=>Macros=> Record...)

Stoyan

2012/10/26 Natalia Chacon <[hidden email]>

--
Dr. Stoyan P. Pavlov, MD, PhD
Departament of Anatomy, Histology and Embryology
Medical University "Prof. Dr. Paraskev Stoyanov", Varna
Prof. Marin Drinov Str.55
9002 Varna
Bulgaria
Tel: +359 (0) 52 - 677 - 086
e-mail: [hidden email]

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

NatyC

Re: Saturation and Quantifying

Dear Stoyan,

Thanks for telling me about the oversaturation.
It became very apparent to me when I began meddling with my images!
The information you've provide to me will greatly assist me.

Thanks a ton,
Natalia

On 10/26/12, Stoyan Pavlov <[hidden email]> wrote:

> Dear Natalia,
> As these are confocal images you are talking about, it is wrong to meddle
> with the color balance! What you need is to separate the channels
> (Image=>Color=>Split channels) and then chose the channel that has your
> protein. (If on other hand these aren't multiphoton images (only one marker
> is recorded) you may skip this step and just convert the image to either
> 16-bit or 8-bit grey (If your color images are 48bit RGB, then chose 16bit
> grey).
> If you are unable to see well the recorded protein you may correct
> brightness and contrast, but be careful not to oversaturate the image. In
> other words your upper brightness limit (the white point) must be set above
> your brightest recorded levels or you will lose information.
> Next step is to segment the Image (i.e. Threshold it) so you get a binary
> image of "protein" and "background".
> Next you can easy calculate occupied surface fraction. (Analyze=>Set
> measurements=> check Area and Limit to Threshold boxes, and then run
> Measure). When you are familiar with all the steps you can easily record
> your own macro (Plugins=>Macros=> Record...)
>
> Stoyan
>
>
>
> 2012/10/26 Natalia Chacon <[hidden email]>
>
>> Dear Leoncio
>>
>> The images I'm working with were provided from my mentor, who stated
>> that the sample had been stained to located a specific protein.
>> From what I understood the images given to me were taken under a
>> confocal microscope.
>>
>> My goal is to quantify the amount of this protein by working off of
>> "how much of the my image is stained “.
>>
>> Whenever the image is edited to draw out the color, the ending product
>> makes it difficult to comprehend how much of the protein is in my
>> sample.
>> (To draw out the color I'm doing Image>Adjust>Color Balance. This is
>> making the staining visible to my eye.)
>> I'm relying on ImageJ to quantify the amount of color there is for the
>> stain.
>> This will give me an idea of how much of the protein is in my image
>>
>> So I'm not looking for the optical density of this image but rather
>> the amount of specific color.
>>
>> Thanks,
>> Natalia
>>
>> On 10/25/12, Vergara, Leoncio A. <[hidden email]> wrote:
>> > I think there is an important component to the question that has not
>> > been
>> > answered and is not clear,
>> > On the original message she wrote: "I was wondering if ImageJ had a way
>> of
>> > detecting colors that aren’t visible to our eye, but are a part of the
>> > image. (Since I'm working with a stain, it’s not present unless its
>> under a
>> > fluorescent light.)"
>> >
>> > What is not clear is how the images are captured, if this is a
>> fluorescent
>> > stain, then the images should not be handled as if they were simple
>> > color
>> > images. I guess there would be filter sets involved and then each
>> > fluorescent channel would be an independent grayscale image, much
>> > easier
>> to
>> > quantify... in terms of area and "amount" (then is not clear what she
>> means
>> > by "amount", intensity? Optical density?
>> >
>> > Leoncio
>> >
>> >
>> > -----Original Message-----
>> > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
>> Vivek
>> > Saraswat
>> > Sent: Thursday, October 25, 2012 4:09 PM
>> > To: [hidden email]
>> > Subject: Re: Saturation and Quantifying
>> >
>> > Hi Natalia,
>> >
>> > So here's the solution:
>> >
>> > 1. In Image J, open your image and then run 8-bit colour mode. It would
>> ask
>> > how many colours you want to digitize the image into? Here you would
>> have to
>> > use your approximation. Like, say 8 colours.
>> >
>> > 2. The RGB values into which the image has been digitized can be viewed
>> from
>> > the LUT (Look up table). Image-->Colour-->Show LUT. Save these RGB
>> > values
>> >
>> > 3. Image-->Type-->RGB Colour. Make as many duplicates of this image as
>> no.
>> > of colours into which image is digitized.
>> >
>> > 4.Then we go to Image->Adjust->Threshold. In threshold colour dialogue
>> box,
>> > set Thresholding method as “Default”, Thresholding Colour as “B&W”
>> > (Black
>> > and White) and Colour Space as “RGB”. Set Colour thresholds on RGB
>> > values
>> > equal to their corresponding values from LUT table.
>> >
>> > 5. Now make each of the images binary and then analyze to get the data
>> about
>> > no. of particles of each colour, size, feret etc.
>> >
>> > I know this may sound a bit confusing so you can write back if anyrhing
>> is
>> > not clear. To summarize, our aim here is to simplify the quantification
>> > process so we first make a guess about no. of colours in which we want
>> > to
>> > digitize it. Then in 8-bit color mode, we can specify how many colours?
>> And
>> > then we find up their RGB values from LUT Table. We then convert the
>> image
>> > back to RGB so that we can again threshold it. Before thresholding we
>> make
>> > no. of copies of this RGB image=No. of colours. Then for each image, we
>> > threshold it with corresponding value from LUT Table, then binarize
>> > each
>> of
>> > them and analyze them.
>> >
>> > If you want to repeat this process a number of times, you have to write
>> > a
>> > macro code!.
>> >
>> > All the best!
>> >
>> > Regards
>> > *Vivek Saraswat*
>> > *Pre-final year Undergraduate*
>> > *Department of Materials Science and Engineering* *IIT Bombay (IIT B),
>> > India* ☎ *+91-9969924136* [hidden email]
>> > http://in.linkedin.com/in/viveksaraswat26<
>> http://home.iitb.ac.in/%7Evivek_saraswat>
>> >
>> >
>> >
>> > On Fri, Oct 26, 2012 at 2:22 AM, Natalia Chacon
>> > <[hidden email]>wrote:
>> >
>> >> Hi Vivek,
>> >>
>> >> Yes, I'm trying to find the area and the amount that a certain colour
>> >> occupies in my whole image and then in each section. (I'm viewing an
>> >> eye with that is stained to locate a protein.)
>> >>
>> >> If there is an algorithm that can assist me with this process I'm up
>> >> for trying it!
>> >>
>> >> Thanks,
>> >> Natalia
>> >>
>> >> On 10/25/12, Vivek Saraswat <[hidden email]> wrote:
>> >> > Hi Natalia,
>> >> >
>> >> > From what I understand from your question is that:
>> >> >
>> >> > You have some sort of a image having different colours blue green
>> >> > red orange of different shades and you want to find out that area
>> >> > fraction
>> >> does
>> >> > each colour occupy?
>> >> >
>> >> > If I am right, do write back, I can help you with an algorithm.
>> >> >
>> >> > Regards
>> >> > *Vivek Saraswat*
>> >> > *Pre-final year Undergraduate*
>> >> > *Department of Materials Science and Engineering* *IIT Bombay (IIT
>> >> > B), India* ☎ *+91-9969924136* [hidden email]
>> >> > http://in.linkedin.com/in/viveksaraswat26<
>> >> http://home.iitb.ac.in/%7Evivek_saraswat>
>> >> >
>> >> >
>> >> >
>> >> > On Thu, Oct 25, 2012 at 11:56 PM, Natalia Chacon
>> >> > <[hidden email]
>> >> >> wrote:
>> >> >
>> >> >> Hello everyone!
>> >> >>
>> >> >> I'm working on a project that has me quantifying a specific color
>> >> >> in the image.
>> >> >> I was wondering if ImageJ had a way of detecting colors that aren’t
>> >> >> visible to our eye, but are a part of the image.
>> >> >> (Since I'm working with a stain, it’s not present unless its under
>> >> >> a fluorescent light.) Also, does anyone have recommendations on how
>> >> >> to quantify colors in using ImageJ?
>> >> >>
>> >> >> Thanks,
>> >> >> Natalia
>> >> >>
>> >> >> --
>> >> >> 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
>> >>
>> >
>> > --
>> > 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
>>
>
>
>
> --
> Dr. Stoyan P. Pavlov, MD, PhD
> Departament of Anatomy, Histology and Embryology
> Medical University "Prof. Dr. Paraskev Stoyanov", Varna
> Prof. Marin Drinov Str.55
> 9002 Varna
> Bulgaria
> Tel: +359 (0) 52 - 677 - 086
> e-mail: [hidden email]
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

NatyC

Re: Saturation and Quantifying

In reply to this post by Christian Goosmann

Dear Christian,

I will certainly try that.

Thanks,
Natalia

On 10/26/12, Christian Goosmann <[hidden email]> wrote:

> Natalia Chacon wrote:
>> Hello everyone!
>>
>> I'm working on a project that has me quantifying a specific color in the
>> image.
>> I was wondering if ImageJ had a way of detecting colors that aren’t
>> visible to our eye, but are a part of the image.
>> (Since I'm working with a stain, it’s not present unless its under a
>> fluorescent light.)
>> Also, does anyone have recommendations on how to quantify colors in
>> using ImageJ?
>>
>> Thanks,
>> Natalia
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
> Dear Natalia,
> if you are speaking of digital images in RGB format, you might want to
> explore 'Image>Adjust>Color Threshold...' from the ImageJ menu. There
> you can filter out those parts of your image that have a certain color,
> saturation, brightness. After segmenting your image with this tool, you
> can analyze the areas in terms of number, size etc.
> Good luck
> Christian
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

slavena trifonova-2

Cancellation of membership

In reply to this post by Stoyan Pavlov

I would like to be deleted from the mailing list PLEASE

________________________________
From: Stoyan Pavlov <[hidden email]>
To: [hidden email]
Sent: Friday, October 26, 2012 8:55 PM
Subject: Re: Saturation and Quantifying

Dear Natalia,
As these are confocal images you are talking about, it is wrong to meddle
with the color balance! What you need is to separate the channels
(Image=>Color=>Split channels) and then chose the channel that has your
protein. (If on other hand these aren't multiphoton images (only one marker
is recorded) you may skip this step and just convert the image to either
16-bit or 8-bit grey (If your color images are 48bit RGB, then chose 16bit
grey).
If you are unable to see well the recorded protein you may correct
brightness and contrast, but be careful not to oversaturate the image. In
other words your upper brightness limit (the white point) must be set above
your brightest recorded levels or you will lose information.
Next step is to segment the Image (i.e. Threshold it) so you get a binary
image of "protein" and "background".
Next you can easy calculate occupied surface fraction. (Analyze=>Set
measurements=> check Area and Limit to Threshold boxes, and then run
Measure). When you are familiar with all the steps you can easily record
your own macro (Plugins=>Macros=> Record...)

Stoyan

2012/10/26 Natalia Chacon <[hidden email]>