ImageJ "mean" vs Photoshop "mean"?

[David Palermo]

I am using my scanner as a densitometer and I am measuring gray patches from a step-wedge in both Photoshop and ImageJ. The mean values are different between each program. Why is that? ImageJ's mean is 41726 and Photoshop's mean of the same patch from the same file is 20861.

Thank you!

David

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

[Saalfeld, Stephan]

Post the patch?

On Sat, 2020-11-28 at 13:26 -0500, David Palermo wrote:

> I am using my scanner as a densitometer and I am measuring gray
> patches from a step-wedge in both Photoshop and ImageJ. The mean
> values are different between each program. Why is that? ImageJ's mean
> is 41726 and Photoshop's mean of the same patch from the same file is
> 20861.
>
> Thank you!
>
> David
>
> --
> ImageJ mailing list:
> https://urldefense.com/v3/__http://imagej.nih.gov/ij/list.html__;!!Eh6p8Q!WiX2hdhTXaPkbqQs1B-O_cXIg7VI21Kut7X8jLZUGNtEbBwU9kFUZOB_vu1vx83uPgu1$
>  

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

[G. Esteban Fernandez]

I’m running an old version of Photoshop (CS2) and it reports the same. Must
be some difference in your workflows. Please post an example image and also
your workflow for each program.

-Esteban


On Sat, Nov 28, 2020 at 12:35 PM Saalfeld, Stephan <
[hidden email]> wrote:

> Post the patch?
>
> On Sat, 2020-11-28 at 13:26 -0500, David Palermo wrote:
> > I am using my scanner as a densitometer and I am measuring gray
> > patches from a step-wedge in both Photoshop and ImageJ. The mean
> > values are different between each program. Why is that? ImageJ's mean
> > is 41726 and Photoshop's mean of the same patch from the same file is
> > 20861.
> >
> > Thank you!
> >
> > David
> >
> > --
> > ImageJ mailing list:
> >
> https://urldefense.com/v3/__http://imagej.nih.gov/ij/list.html__;!!Eh6p8Q!WiX2hdhTXaPkbqQs1B-O_cXIg7VI21Kut7X8jLZUGNtEbBwU9kFUZOB_vu1vx83uPgu1$
> >
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

[GRADD CO]

Simple observation...for what it's worth...your ImageJ mean is nearly
exactly double that of your Photoshop mean.
REZA KARAMOOZ

CEO, GRADD
<https://mailtrack.io/trace/link/7ad9a122c52d274bf1f241c3c5053637b5231553?url=https%3A%2F%2Fgradd.co&userId=5893829&signature=510720a1f91ee2f4>
Civic Leader | USAF AMMOS
<https://mailtrack.io/trace/link/bc8e73ac6322511137801c139ea32ad005963ab1?url=https%3A%2F%2Fwww.nellis.af.mil%2FUnits%2F57-WG%2F&userId=5893829&signature=03184a6d9a9c4ed3>,
57th Wing, Nellis AFB
7028799100 | [hidden email]
2900 Meade Ave Ste 8
Las Vegas, NV 89102
<https://mailtrack.io/trace/link/8dd2a7f4740dd769a223a120bf2ad5e7adee8378?url=https%3A%2F%2Fwww.linkedin.com%2Fin%2Fgradd%2F&userId=5893829&signature=6a09150adafa35a0>
  Click to schedule a meeting
<https://mailtrack.io/trace/link/449bd34691b16bdf68f2ab55ac5132dae98ac667?url=https%3A%2F%2Fgradd.as.me%2Fschedule.php&userId=5893829&signature=d562ff6e29b072ff>

CONFIDENTIAL COMMUNICATION: This email message and any attachment may
contain privileged and confidential information intended only for the use
of the individual or entity to which the email is addressed. If the reader
of this message is not the intended recipient or the employee or agent
responsible to deliver it to the intended recipient, that person is hereby
notified that any dissemination, distribution or copying of this
communication is prohibited.


On Sat, Nov 28, 2020 at 4:39 PM G. Esteban Fernandez <
[hidden email]> wrote:

> I’m running an old version of Photoshop (CS2) and it reports the same. Must
> be some difference in your workflows. Please post an example image and also
> your workflow for each program.
>
> -Esteban
>
>
> On Sat, Nov 28, 2020 at 12:35 PM Saalfeld, Stephan <
> [hidden email]> wrote:
>
> > Post the patch?
> >
> > On Sat, 2020-11-28 at 13:26 -0500, David Palermo wrote:
> > > I am using my scanner as a densitometer and I am measuring gray
> > > patches from a step-wedge in both Photoshop and ImageJ. The mean
> > > values are different between each program. Why is that? ImageJ's mean
> > > is 41726 and Photoshop's mean of the same patch from the same file is
> > > 20861.
> > >
> > > Thank you!
> > >
> > > David
> > >
> > > --
> > > ImageJ mailing list:
> > >
> >
> https://urldefense.com/v3/__http://imagej.nih.gov/ij/list.html__;!!Eh6p8Q!WiX2hdhTXaPkbqQs1B-O_cXIg7VI21Kut7X8jLZUGNtEbBwU9kFUZOB_vu1vx83uPgu1$
> <https://mailtrack.io/trace/link/04b6912e0be97160f39cc1ecc5d99a311d22de36?url=https%3A%2F%2Furldefense.com%2Fv3%2F__http%3A%2F%2Fimagej.nih.gov%2Fij%2Flist.html__%3B!!Eh6p8Q!WiX2hdhTXaPkbqQs1B-O_cXIg7VI21Kut7X8jLZUGNtEbBwU9kFUZOB_vu1vx83uPgu1%24&userId=5893829&signature=3f2493d71e8ba57f>
> > >
> >
> >
> > --
> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> <https://mailtrack.io/trace/link/9842e3cf9460dc8bbbd6a223188cf379826229e6?url=http%3A%2F%2Fimagej.nih.gov%2Fij%2Flist.html&userId=5893829&signature=ec770f6ef2ac1ec5>
> >
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> <https://mailtrack.io/trace/link/672b393203ec982ee518b42d8a91e1bf9355ce6f?url=http%3A%2F%2Fimagej.nih.gov%2Fij%2Flist.html&userId=5893829&signature=02b5eed4ee4c63dd>
>

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

[Herbie]

Greetings David,

be careful with Photoshop. It was/is not designed for scientific image
processing and if you don't carefully control all options, you will end
up with results as the reported one. One reason for the different
results may be due to gamma-correction in photoshop but there are many
other possible reasons.

For scientific image processing and analyses always use a dedicated
software, e.g. ImageJ.

Regards

Herbie

:::::::::::::::::::::::::::::::::::::::::::
Am 28.11.20 um 19:26 schrieb David Palermo:
> I am using my scanner as a densitometer and I am measuring gray patches from a step-wedge in both Photoshop and ImageJ. The mean values are different between each program. Why is that? ImageJ's mean is 41726 and Photoshop's mean of the same patch from the same file is 20861.
>
> Thank you!
>
> David
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

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

[Michael Schmid]

Hi David,

if your image file has tiff format, it could have some nonlinear
transfer function (e.g. a color calibration function) that is taken into
account by photoshop, not by ImageJ. ImageJ is usually used for images
from scientific cameras that have linear transfer functions, so it does
not take the tiff tags for nonlinear transfer functions into account
(tag decimal 301 = hex 012D "TransferFunction", 50940 = hex C6FC
"ProfileToneCurve")

If you switch on debug mode (under Edit>Options>Misc), you will see
whether such a tiff tag is present. As many others have said already, it
would have been easier with a sample file.

Michael
________________________________________________________________
On 28.11.20 19:26, David Palermo wrote:
> I am using my scanner as a densitometer and I am measuring gray patches from a step-wedge in both Photoshop and ImageJ. The mean values are different between each program. Why is that? ImageJ's mean is 41726 and Photoshop's mean of the same patch from the same file is 20861.
>
> Thank you!
>
> David

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

[Kenneth Sloan-2]

A meta-comment.  I am a bit put off by the  characterization of devices which use non-linear transfer functions as "non-scientific".  

One almost gets the impression that the fact that ImageJ ignores transfer function information present in a .tif file is somehow a "feature".  I consider it a "limitation", if not actually a "bug".

If the image source happens to use a non-linear transfer function, I would consider it "unscientific" to ignore it.  I consider it only a little bit less bad to say "oh, that image source is 'unscientific' - so we won't support it".  More accurate is to say "ImageJ is limited to linear images".

Actually, upon reflection, I *do* consider it a "bug" that ImageJ loads a .tif image with a non-linear transform noted - and instead displays it linearly without even a warning (if, in fact, that's what it's doing).

--
Kenneth Sloan
[hidden email]
Vision is the art of seeing what is invisible to others.

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

[Robert Smith]

I have to agree with this statement. The more options for images, the more ‘scientific’ it becomes due  to the variety of ways an image has to be captured to obtain the data required.
Robert J. Smith

Sent from Mail<https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10

From: Kenneth Sloan<mailto:[hidden email]>
Sent: Monday, November 30, 2020 8:03 PM
To: [hidden email]<mailto:[hidden email]>
Subject: Re: ImageJ "mean" vs Photoshop "mean"?

A meta-comment.  I am a bit put off by the  characterization of devices which use non-linear transfer functions as "non-scientific".

One almost gets the impression that the fact that ImageJ ignores transfer function information present in a .tif file is somehow a "feature".  I consider it a "limitation", if not actually a "bug".

If the image source happens to use a non-linear transfer function, I would consider it "unscientific" to ignore it.  I consider it only a little bit less bad to say "oh, that image source is 'unscientific' - so we won't support it".  More accurate is to say "ImageJ is limited to linear images".

Actually, upon reflection, I *do* consider it a "bug" that ImageJ loads a .tif image with a non-linear transform noted - and instead displays it linearly without even a warning (if, in fact, that's what it's doing).

--
Kenneth Sloan
[hidden email]
Vision is the art of seeing what is invisible to others.

--
ImageJ mailing list: https://nam10.safelinks.protection.outlook.com/?url=http%3A%2F%2Fimagej.nih.gov%2Fij%2Flist.html&amp;data=04%7C01%7C%7C21fcc0730c12484c147608d89594f04c%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637423814195366663%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp;sdata=JAKlU7npgh7dnvhQ8F1VgZJadkRg55nFN81ikkN7eXc%3D&amp;reserved=0


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