Image resolution versus Screen resolution 
Author Message
 Image resolution versus Screen resolution

I'm writing a small MDI app to measure the size of features in TIFF (or
others) images that are loaded in pictureboxes on child forms. Currently the
code captures these distances in pixels so to convert to a physical
dimension some form of resolution (DpiX/DpiY) information will be necessary.

The question in my mind right now is, which resolution applies? The screen
resolution, or the actual image resolution? Both can be readily obtained in
code.

My guess is that no matter whether the image resolution is higher or lower
than the screen resolution, it is the screen resolution that applies in this
case.

Is there more behind this than meets the eye....?

TIA



Sat, 30 Apr 2005 23:35:00 GMT  
 Image resolution versus Screen resolution
The physical dimensions of a image depend on what you are planning on doing
with it. The pixel dimensions are absolute. Some image formats carry dpi
information aroung with them, bit you are more or less free to do whatever
you want with the pixels.

If, for example, the images are aerial photographs, you should have the size
of the pixel available (e.g. 6m, 10m, etc.) which you can use to convert to
physical distances. If you scanned the images, hopefully you recorded the
scan settings, and you would use those to get the physical size. In you are
actually looking for the physical dimensions of the image on the monitor you
need the dpi of the monitor.

If you provide specifics of exactly what you are doing it would be easier to
give you a definite answer.

Colin


Quote:
> I'm writing a small MDI app to measure the size of features in TIFF (or
> others) images that are loaded in pictureboxes on child forms. Currently
the
> code captures these distances in pixels so to convert to a physical
> dimension some form of resolution (DpiX/DpiY) information will be
necessary.

> The question in my mind right now is, which resolution applies? The screen
> resolution, or the actual image resolution? Both can be readily obtained
in
> code.

> My guess is that no matter whether the image resolution is higher or lower
> than the screen resolution, it is the screen resolution that applies in
this
> case.

> Is there more behind this than meets the eye....?

> TIA



Sun, 01 May 2005 02:30:10 GMT  
 Image resolution versus Screen resolution
It's always physical resolution. That's what the camera captured. I'll have
to think on this one overnight to figure out how to back out physical
dimensions.

Colin


Quote:
> I'm sorry Colin

> The images are taken using a high resolution CCD camera attached to a Lab
> microscope. Although the images are magnified by the microsocope, I know
the
> magnification that is used. Using the camera software I can save the
images
> to any size or resolution I desire....well as much as is allowed by the
> image format I choose. For example I chose to save them in a TIFF format
at
> 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose this format
> because I wanted to capture as much detail as possible. The files are
about
> 6 MB! I did not want to bore people with those details but if they're
> necessary for a good answer...no problem.

> I'm not sure how to determine the resolution of these image files. I was
> hoping to use C# to do this for me (DpiX & Y properties) but I'm still
> wrestling with "how to do this in an MDI Child form" in other posts on
this
> forum under my name. Regardless, when I view these images in my program, I
> believe that I am looking at them at my monitor's resolution which I have
> set to 120 DPI.

> Ignoring the real/apparent magnification effects of capturing the images
> themselves, when I measure something in an image that is 100 pixels in
> length is it 100 pixels divided by the screen resolution, or 100 pixels
> divided by the image's resolution to get at the physical length?

> TIA



> > The physical dimensions of a image depend on what you are planning on
> doing
> > with it. The pixel dimensions are absolute. Some image formats carry dpi
> > information aroung with them, bit you are more or less free to do
whatever
> > you want with the pixels.

> > If, for example, the images are aerial photographs, you should have the
> size
> > of the pixel available (e.g. 6m, 10m, etc.) which you can use to convert
> to
> > physical distances. If you scanned the images, hopefully you recorded
the
> > scan settings, and you would use those to get the physical size. In you
> are
> > actually looking for the physical dimensions of the image on the monitor
> you
> > need the dpi of the monitor.

> > If you provide specifics of exactly what you are doing it would be
easier
> to
> > give you a definite answer.

> > Colin



> > > I'm writing a small MDI app to measure the size of features in TIFF
(or
> > > others) images that are loaded in pictureboxes on child forms.
Currently
> > the
> > > code captures these distances in pixels so to convert to a physical
> > > dimension some form of resolution (DpiX/DpiY) information will be
> > necessary.

> > > The question in my mind right now is, which resolution applies? The
> screen
> > > resolution, or the actual image resolution? Both can be readily
obtained
> > in
> > > code.

> > > My guess is that no matter whether the image resolution is higher or
> lower
> > > than the screen resolution, it is the screen resolution that applies
in
> > this
> > > case.

> > > Is there more behind this than meets the eye....?

> > > TIA



Sun, 01 May 2005 05:58:33 GMT  
 Image resolution versus Screen resolution
I'm sorry Colin

The images are taken using a high resolution CCD camera attached to a Lab
microscope. Although the images are magnified by the microsocope, I know the
magnification that is used. Using the camera software I can save the images
to any size or resolution I desire....well as much as is allowed by the
image format I choose. For example I chose to save them in a TIFF format at
1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose this format
because I wanted to capture as much detail as possible. The files are about
6 MB! I did not want to bore people with those details but if they're
necessary for a good answer...no problem.

I'm not sure how to determine the resolution of these image files. I was
hoping to use C# to do this for me (DpiX & Y properties) but I'm still
wrestling with "how to do this in an MDI Child form" in other posts on this
forum under my name. Regardless, when I view these images in my program, I
believe that I am looking at them at my monitor's resolution which I have
set to 120 DPI.

Ignoring the real/apparent magnification effects of capturing the images
themselves, when I measure something in an image that is 100 pixels in
length is it 100 pixels divided by the screen resolution, or 100 pixels
divided by the image's resolution to get at the physical length?

TIA


Quote:
> The physical dimensions of a image depend on what you are planning on
doing
> with it. The pixel dimensions are absolute. Some image formats carry dpi
> information aroung with them, bit you are more or less free to do whatever
> you want with the pixels.

> If, for example, the images are aerial photographs, you should have the
size
> of the pixel available (e.g. 6m, 10m, etc.) which you can use to convert
to
> physical distances. If you scanned the images, hopefully you recorded the
> scan settings, and you would use those to get the physical size. In you
are
> actually looking for the physical dimensions of the image on the monitor
you
> need the dpi of the monitor.

> If you provide specifics of exactly what you are doing it would be easier
to
> give you a definite answer.

> Colin



> > I'm writing a small MDI app to measure the size of features in TIFF (or
> > others) images that are loaded in pictureboxes on child forms. Currently
> the
> > code captures these distances in pixels so to convert to a physical
> > dimension some form of resolution (DpiX/DpiY) information will be
> necessary.

> > The question in my mind right now is, which resolution applies? The
screen
> > resolution, or the actual image resolution? Both can be readily obtained
> in
> > code.

> > My guess is that no matter whether the image resolution is higher or
lower
> > than the screen resolution, it is the screen resolution that applies in
> this
> > case.

> > Is there more behind this than meets the eye....?

> > TIA



Sun, 01 May 2005 03:03:48 GMT  
 Image resolution versus Screen resolution
Colin

When I say "measure", I mean I am looking at an image on my monitor and
using the Mouse events to capture X-Y positions which give me the distance
the Mouse traveled in units of Pixels. I am not physically measuring with a
"ruler" or anything like that....I should have mentioned.

From the Mouse distance in pixels I think I can divide by the screen
resolution to get a physical size but I'm not sure.....

Gerry


Quote:
> It's always physical resolution. That's what the camera captured. I'll
have
> to think on this one overnight to figure out how to back out physical
> dimensions.

> Colin



> > I'm sorry Colin

> > The images are taken using a high resolution CCD camera attached to a
Lab
> > microscope. Although the images are magnified by the microsocope, I know
> the
> > magnification that is used. Using the camera software I can save the
> images
> > to any size or resolution I desire....well as much as is allowed by the
> > image format I choose. For example I chose to save them in a TIFF format
> at
> > 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose this
format
> > because I wanted to capture as much detail as possible. The files are
> about
> > 6 MB! I did not want to bore people with those details but if they're
> > necessary for a good answer...no problem.

> > I'm not sure how to determine the resolution of these image files. I was
> > hoping to use C# to do this for me (DpiX & Y properties) but I'm still
> > wrestling with "how to do this in an MDI Child form" in other posts on
> this
> > forum under my name. Regardless, when I view these images in my program,
I
> > believe that I am looking at them at my monitor's resolution which I
have
> > set to 120 DPI.

> > Ignoring the real/apparent magnification effects of capturing the images
> > themselves, when I measure something in an image that is 100 pixels in
> > length is it 100 pixels divided by the screen resolution, or 100 pixels
> > divided by the image's resolution to get at the physical length?

> > TIA



> > > The physical dimensions of a image depend on what you are planning on
> > doing
> > > with it. The pixel dimensions are absolute. Some image formats carry
dpi
> > > information aroung with them, bit you are more or less free to do
> whatever
> > > you want with the pixels.

> > > If, for example, the images are aerial photographs, you should have
the
> > size
> > > of the pixel available (e.g. 6m, 10m, etc.) which you can use to
convert
> > to
> > > physical distances. If you scanned the images, hopefully you recorded
> the
> > > scan settings, and you would use those to get the physical size. In
you
> > are
> > > actually looking for the physical dimensions of the image on the
monitor
> > you
> > > need the dpi of the monitor.

> > > If you provide specifics of exactly what you are doing it would be
> easier
> > to
> > > give you a definite answer.

> > > Colin



> > > > I'm writing a small MDI app to measure the size of features in TIFF
> (or
> > > > others) images that are loaded in pictureboxes on child forms.
> Currently
> > > the
> > > > code captures these distances in pixels so to convert to a physical
> > > > dimension some form of resolution (DpiX/DpiY) information will be
> > > necessary.

> > > > The question in my mind right now is, which resolution applies? The
> > screen
> > > > resolution, or the actual image resolution? Both can be readily
> obtained
> > > in
> > > > code.

> > > > My guess is that no matter whether the image resolution is higher or
> > lower
> > > > than the screen resolution, it is the screen resolution that applies
> in
> > > this
> > > > case.

> > > > Is there more behind this than meets the eye....?

> > > > TIA



Sun, 01 May 2005 10:14:05 GMT  
 Image resolution versus Screen resolution
I did actually understand that you are trying to extract the physical
dimensions of an object in your image.

I think the easiest and safest way is to put an object of known dimensions
in the microscope, photograph it and then calculate the pixel size from
that.

You don't need any information about the monitor resolution. The only thing
that matters is the number of pixels. Some screens use 72dpi, others use
96dpi, and I don't think those numbers are precise (e.g. a 15in monitor at
VGA resolution will have a different dpi than a 17in monitor at the same
resolution).

e.g. if a 1mm feature is magnified 100X, it will "appear" to be 100mm. I'm
assuming that the camera has some optics and the microscope is not
projecting directly onto the CCD. Assuming the camera can capture the full
100mm image, and for argument's sake let's assume that it covers 1000 of the
1600 pixels, that means a 1mm feature maps to 1000 pixels, or 1 pixel
represents 1 micrometre.

Does that make sense?

Colin


Quote:
> Colin

> When I say "measure", I mean I am looking at an image on my monitor and
> using the Mouse events to capture X-Y positions which give me the distance
> the Mouse traveled in units of Pixels. I am not physically measuring with
a
> "ruler" or anything like that....I should have mentioned.

> From the Mouse distance in pixels I think I can divide by the screen
> resolution to get a physical size but I'm not sure.....

> Gerry



> > It's always physical resolution. That's what the camera captured. I'll
> have
> > to think on this one overnight to figure out how to back out physical
> > dimensions.

> > Colin



> > > I'm sorry Colin

> > > The images are taken using a high resolution CCD camera attached to a
> Lab
> > > microscope. Although the images are magnified by the microsocope, I
know
> > the
> > > magnification that is used. Using the camera software I can save the
> > images
> > > to any size or resolution I desire....well as much as is allowed by
the
> > > image format I choose. For example I chose to save them in a TIFF
format
> > at
> > > 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose this
> format
> > > because I wanted to capture as much detail as possible. The files are
> > about
> > > 6 MB! I did not want to bore people with those details but if they're
> > > necessary for a good answer...no problem.

> > > I'm not sure how to determine the resolution of these image files. I
was
> > > hoping to use C# to do this for me (DpiX & Y properties) but I'm still
> > > wrestling with "how to do this in an MDI Child form" in other posts on
> > this
> > > forum under my name. Regardless, when I view these images in my
program,
> I
> > > believe that I am looking at them at my monitor's resolution which I
> have
> > > set to 120 DPI.

> > > Ignoring the real/apparent magnification effects of capturing the
images
> > > themselves, when I measure something in an image that is 100 pixels in
> > > length is it 100 pixels divided by the screen resolution, or 100
pixels
> > > divided by the image's resolution to get at the physical length?

> > > TIA



> > > > The physical dimensions of a image depend on what you are planning
on
> > > doing
> > > > with it. The pixel dimensions are absolute. Some image formats carry
> dpi
> > > > information aroung with them, bit you are more or less free to do
> > whatever
> > > > you want with the pixels.

> > > > If, for example, the images are aerial photographs, you should have
> the
> > > size
> > > > of the pixel available (e.g. 6m, 10m, etc.) which you can use to
> convert
> > > to
> > > > physical distances. If you scanned the images, hopefully you
recorded
> > the
> > > > scan settings, and you would use those to get the physical size. In
> you
> > > are
> > > > actually looking for the physical dimensions of the image on the
> monitor
> > > you
> > > > need the dpi of the monitor.

> > > > If you provide specifics of exactly what you are doing it would be
> > easier
> > > to
> > > > give you a definite answer.

> > > > Colin



> > > > > I'm writing a small MDI app to measure the size of features in
TIFF
> > (or
> > > > > others) images that are loaded in pictureboxes on child forms.
> > Currently
> > > > the
> > > > > code captures these distances in pixels so to convert to a
physical
> > > > > dimension some form of resolution (DpiX/DpiY) information will be
> > > > necessary.

> > > > > The question in my mind right now is, which resolution applies?
The
> > > screen
> > > > > resolution, or the actual image resolution? Both can be readily
> > obtained
> > > > in
> > > > > code.

> > > > > My guess is that no matter whether the image resolution is higher
or
> > > lower
> > > > > than the screen resolution, it is the screen resolution that
applies
> > in
> > > > this
> > > > > case.

> > > > > Is there more behind this than meets the eye....?

> > > > > TIA



Sun, 01 May 2005 23:23:55 GMT  
 Image resolution versus Screen resolution
Colin

I knew there was a good reason to pursue this here....! Thank you very much
for{*filter*} in and being patient....

I have EXACTLY what you suggest. These images I was referring to are camera
shots of what is known as a "Stage Micrometer". Just in case, a stage
micrometer is a precisely ruled reference (traceable to US national Bureau
of Standard) necessary to calibrate equipment used to make measurements at
the microscope itself. I want to measure camera shots.... eye fatigue you
know!

In this case, the "rulings" on the stage micrometer are in 0.0005"
increments. I apologize for US units, I promise to take pictures of the mm
scaled Stage Micrometer!!!!

If I understand you correctly:

1) The monitor resolution is not involved. I am only concerned with the
magnification of the actual image and
    the number of pixels it represents on the screen.

2) At 100X, one graduation mark of 0.0005" will "appear" to be the same size
as  an 0.05" image taken at 1:1.

3) If the distance for this 0.0005" graduation mark in the image is 100
pixels (by the Mouse movement), then the pixel size is
    0.0005" / 100 pixels  or  0.00005" per pixel.

Now for my extrapolations:

1) The pixel size above is ONLY good for measuring other images taken at
100X, I would have to perform similar calibration for
    other magnifications?

2) I can only measure items in images that have known magnifications (and
physical calibrations)? I may not always have this "luxury". Are there
techniques for measuring "unknowns"?

3) As long as the complete item to be measured is visible in the image (and
on the screen) the screen size (1024 by 768 and others) and the image size
(1600 by 1200 and others) are not involved.

TIA and I owe you!

Gerry


Quote:
> I did actually understand that you are trying to extract the physical
> dimensions of an object in your image.

> I think the easiest and safest way is to put an object of known dimensions
> in the microscope, photograph it and then calculate the pixel size from
> that.

> You don't need any information about the monitor resolution. The only
thing
> that matters is the number of pixels. Some screens use 72dpi, others use
> 96dpi, and I don't think those numbers are precise (e.g. a 15in monitor at
> VGA resolution will have a different dpi than a 17in monitor at the same
> resolution).

> e.g. if a 1mm feature is magnified 100X, it will "appear" to be 100mm. I'm
> assuming that the camera has some optics and the microscope is not
> projecting directly onto the CCD. Assuming the camera can capture the full
> 100mm image, and for argument's sake let's assume that it covers 1000 of
the
> 1600 pixels, that means a 1mm feature maps to 1000 pixels, or 1 pixel
> represents 1 micrometre.

> Does that make sense?

> Colin



> > Colin

> > When I say "measure", I mean I am looking at an image on my monitor and
> > using the Mouse events to capture X-Y positions which give me the
distance
> > the Mouse traveled in units of Pixels. I am not physically measuring
with
> a
> > "ruler" or anything like that....I should have mentioned.

> > From the Mouse distance in pixels I think I can divide by the screen
> > resolution to get a physical size but I'm not sure.....

> > Gerry



> > > It's always physical resolution. That's what the camera captured. I'll
> > have
> > > to think on this one overnight to figure out how to back out physical
> > > dimensions.

> > > Colin



> > > > I'm sorry Colin

> > > > The images are taken using a high resolution CCD camera attached to
a
> > Lab
> > > > microscope. Although the images are magnified by the microsocope, I
> know
> > > the
> > > > magnification that is used. Using the camera software I can save the
> > > images
> > > > to any size or resolution I desire....well as much as is allowed by
> the
> > > > image format I choose. For example I chose to save them in a TIFF
> format
> > > at
> > > > 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose this
> > format
> > > > because I wanted to capture as much detail as possible. The files
are
> > > about
> > > > 6 MB! I did not want to bore people with those details but if
they're
> > > > necessary for a good answer...no problem.

> > > > I'm not sure how to determine the resolution of these image files. I
> was
> > > > hoping to use C# to do this for me (DpiX & Y properties) but I'm
still
> > > > wrestling with "how to do this in an MDI Child form" in other posts
on
> > > this
> > > > forum under my name. Regardless, when I view these images in my
> program,
> > I
> > > > believe that I am looking at them at my monitor's resolution which I
> > have
> > > > set to 120 DPI.

> > > > Ignoring the real/apparent magnification effects of capturing the
> images
> > > > themselves, when I measure something in an image that is 100 pixels
in
> > > > length is it 100 pixels divided by the screen resolution, or 100
> pixels
> > > > divided by the image's resolution to get at the physical length?

> > > > TIA



> > > > > The physical dimensions of a image depend on what you are planning
> on
> > > > doing
> > > > > with it. The pixel dimensions are absolute. Some image formats
carry
> > dpi
> > > > > information aroung with them, bit you are more or less free to do
> > > whatever
> > > > > you want with the pixels.

> > > > > If, for example, the images are aerial photographs, you should
have
> > the
> > > > size
> > > > > of the pixel available (e.g. 6m, 10m, etc.) which you can use to
> > convert
> > > > to
> > > > > physical distances. If you scanned the images, hopefully you
> recorded
> > > the
> > > > > scan settings, and you would use those to get the physical size.
In
> > you
> > > > are
> > > > > actually looking for the physical dimensions of the image on the
> > monitor
> > > > you
> > > > > need the dpi of the monitor.

> > > > > If you provide specifics of exactly what you are doing it would be
> > > easier
> > > > to
> > > > > give you a definite answer.

> > > > > Colin



> > > > > > I'm writing a small MDI app to measure the size of features in
> TIFF
> > > (or
> > > > > > others) images that are loaded in pictureboxes on child forms.
> > > Currently
> > > > > the
> > > > > > code captures these distances in pixels so to convert to a
> physical
> > > > > > dimension some form of resolution (DpiX/DpiY) information will
be
> > > > > necessary.

> > > > > > The question in my mind right now is, which resolution applies?
> The
> > > > screen
> > > > > > resolution, or the actual image resolution? Both can be readily
> > > obtained
> > > > > in
> > > > > > code.

> > > > > > My guess is that no matter whether the image resolution is
higher
> or
> > > > lower
> > > > > > than the screen resolution, it is the screen resolution that
> applies
> > > in
> > > > > this
> > > > > > case.

> > > > > > Is there more behind this than meets the eye....?

> > > > > > TIA



Mon, 02 May 2005 02:48:04 GMT  
 Image resolution versus Screen resolution
I appear to have missed your message for a couple days. Sorry. Too much
traffic in this group.

As for calibration, if the camera is fixed to the microscope, you should
only need to calibrate once at each magnification. Once you have the ratio,
you're all set. If the camera is not permanently attached, do a few
calibrations, removing it in between and changing the magnification to see
if that affects the calibration (and to determine if it remains within an
acceptable error range). Also, use the largest visible known length to
calibrate (the difference between 9 or 10 pixels is 10 times the difference
between 99 and 100 pixels, or 1/10th the error).

You don't actually need to worry about the magnification when doing the
calibration. The only reason it matters is to keep track of the conversion.
3) has it exactly right.

As for unknowns, you need to at least know the magnification or have a
calibration. For example, if at 100X 1px = 1E-3mm and at 200X 1px =
500E-6mm, then it's reasonable to assume that at 150X you are going to be
halfway between (or 750E-6mm/px). If you don't know the magnification or
calibration, then you need to have something in the image of known
dimensions so you can calibrate. There's no way around that.

Colin


Quote:
> Colin

> I knew there was a good reason to pursue this here....! Thank you very
much
> for{*filter*} in and being patient....

> I have EXACTLY what you suggest. These images I was referring to are
camera
> shots of what is known as a "Stage Micrometer". Just in case, a stage
> micrometer is a precisely ruled reference (traceable to US national Bureau
> of Standard) necessary to calibrate equipment used to make measurements at
> the microscope itself. I want to measure camera shots.... eye fatigue you
> know!

> In this case, the "rulings" on the stage micrometer are in 0.0005"
> increments. I apologize for US units, I promise to take pictures of the mm
> scaled Stage Micrometer!!!!

> If I understand you correctly:

> 1) The monitor resolution is not involved. I am only concerned with the
> magnification of the actual image and
>     the number of pixels it represents on the screen.

> 2) At 100X, one graduation mark of 0.0005" will "appear" to be the same
size
> as  an 0.05" image taken at 1:1.

> 3) If the distance for this 0.0005" graduation mark in the image is 100
> pixels (by the Mouse movement), then the pixel size is
>     0.0005" / 100 pixels  or  0.00005" per pixel.

> Now for my extrapolations:

> 1) The pixel size above is ONLY good for measuring other images taken at
> 100X, I would have to perform similar calibration for
>     other magnifications?

> 2) I can only measure items in images that have known magnifications (and
> physical calibrations)? I may not always have this "luxury". Are there
> techniques for measuring "unknowns"?

> 3) As long as the complete item to be measured is visible in the image
(and
> on the screen) the screen size (1024 by 768 and others) and the image size
> (1600 by 1200 and others) are not involved.

> TIA and I owe you!

> Gerry



> > I did actually understand that you are trying to extract the physical
> > dimensions of an object in your image.

> > I think the easiest and safest way is to put an object of known
dimensions
> > in the microscope, photograph it and then calculate the pixel size from
> > that.

> > You don't need any information about the monitor resolution. The only
> thing
> > that matters is the number of pixels. Some screens use 72dpi, others use
> > 96dpi, and I don't think those numbers are precise (e.g. a 15in monitor
at
> > VGA resolution will have a different dpi than a 17in monitor at the same
> > resolution).

> > e.g. if a 1mm feature is magnified 100X, it will "appear" to be 100mm.
I'm
> > assuming that the camera has some optics and the microscope is not
> > projecting directly onto the CCD. Assuming the camera can capture the
full
> > 100mm image, and for argument's sake let's assume that it covers 1000 of
> the
> > 1600 pixels, that means a 1mm feature maps to 1000 pixels, or 1 pixel
> > represents 1 micrometre.

> > Does that make sense?

> > Colin



> > > Colin

> > > When I say "measure", I mean I am looking at an image on my monitor
and
> > > using the Mouse events to capture X-Y positions which give me the
> distance
> > > the Mouse traveled in units of Pixels. I am not physically measuring
> with
> > a
> > > "ruler" or anything like that....I should have mentioned.

> > > From the Mouse distance in pixels I think I can divide by the screen
> > > resolution to get a physical size but I'm not sure.....

> > > Gerry



> > > > It's always physical resolution. That's what the camera captured.
I'll
> > > have
> > > > to think on this one overnight to figure out how to back out
physical
> > > > dimensions.

> > > > Colin



> > > > > I'm sorry Colin

> > > > > The images are taken using a high resolution CCD camera attached
to
> a
> > > Lab
> > > > > microscope. Although the images are magnified by the microsocope,
I
> > know
> > > > the
> > > > > magnification that is used. Using the camera software I can save
the
> > > > images
> > > > > to any size or resolution I desire....well as much as is allowed
by
> > the
> > > > > image format I choose. For example I chose to save them in a TIFF
> > format
> > > > at
> > > > > 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose
this
> > > format
> > > > > because I wanted to capture as much detail as possible. The files
> are
> > > > about
> > > > > 6 MB! I did not want to bore people with those details but if
> they're
> > > > > necessary for a good answer...no problem.

> > > > > I'm not sure how to determine the resolution of these image files.
I
> > was
> > > > > hoping to use C# to do this for me (DpiX & Y properties) but I'm
> still
> > > > > wrestling with "how to do this in an MDI Child form" in other
posts
> on
> > > > this
> > > > > forum under my name. Regardless, when I view these images in my
> > program,
> > > I
> > > > > believe that I am looking at them at my monitor's resolution which
I
> > > have
> > > > > set to 120 DPI.

> > > > > Ignoring the real/apparent magnification effects of capturing the
> > images
> > > > > themselves, when I measure something in an image that is 100
pixels
> in
> > > > > length is it 100 pixels divided by the screen resolution, or 100
> > pixels
> > > > > divided by the image's resolution to get at the physical length?

> > > > > TIA



> > > > > > The physical dimensions of a image depend on what you are
planning
> > on
> > > > > doing
> > > > > > with it. The pixel dimensions are absolute. Some image formats
> carry
> > > dpi
> > > > > > information aroung with them, bit you are more or less free to
do
> > > > whatever
> > > > > > you want with the pixels.

> > > > > > If, for example, the images are aerial photographs, you should
> have
> > > the
> > > > > size
> > > > > > of the pixel available (e.g. 6m, 10m, etc.) which you can use to
> > > convert
> > > > > to
> > > > > > physical distances. If you scanned the images, hopefully you
> > recorded
> > > > the
> > > > > > scan settings, and you would use those to get the physical size.
> In
> > > you
> > > > > are
> > > > > > actually looking for the physical dimensions of the image on the
> > > monitor
> > > > > you
> > > > > > need the dpi of the monitor.

> > > > > > If you provide specifics of exactly what you are doing it would
be
> > > > easier
> > > > > to
> > > > > > give you a definite answer.

> > > > > > Colin



> > > > > > > I'm writing a small MDI app to measure the size of features in
> > TIFF
> > > > (or
> > > > > > > others) images that are loaded in pictureboxes on child forms.
> > > > Currently
> > > > > > the
> > > > > > > code captures these distances in pixels so to convert to a
> > physical
> > > > > > > dimension some form of resolution (DpiX/DpiY) information will
> be
> > > > > > necessary.

> > > > > > > The question in my mind right now is, which resolution
applies?
> > The
> > > > > screen
> > > > > > > resolution, or the actual image resolution? Both can be
readily
> > > > obtained
> > > > > > in
> > > > > > > code.

> > > > > > > My guess is that no matter whether the image resolution is
> higher
> > or
> > > > > lower
> > > > > > > than the screen resolution, it is the screen resolution that
> > applies
> > > > in
> > > > > > this
> > > > > > > case.

> > > > > > > Is there more behind this than meets the eye....?

> > > > > > > TIA



Wed, 04 May 2005 02:37:39 GMT  
 Image resolution versus Screen resolution
Colin

Thanks for getting me up to speed!

/Gerry


Quote:
> I appear to have missed your message for a couple days. Sorry. Too much
> traffic in this group.

> As for calibration, if the camera is fixed to the microscope, you should
> only need to calibrate once at each magnification. Once you have the
ratio,
> you're all set. If the camera is not permanently attached, do a few
> calibrations, removing it in between and changing the magnification to see
> if that affects the calibration (and to determine if it remains within an
> acceptable error range). Also, use the largest visible known length to
> calibrate (the difference between 9 or 10 pixels is 10 times the
difference
> between 99 and 100 pixels, or 1/10th the error).

> You don't actually need to worry about the magnification when doing the
> calibration. The only reason it matters is to keep track of the
conversion.
> 3) has it exactly right.

> As for unknowns, you need to at least know the magnification or have a
> calibration. For example, if at 100X 1px = 1E-3mm and at 200X 1px =
> 500E-6mm, then it's reasonable to assume that at 150X you are going to be
> halfway between (or 750E-6mm/px). If you don't know the magnification or
> calibration, then you need to have something in the image of known
> dimensions so you can calibrate. There's no way around that.

> Colin



> > Colin

> > I knew there was a good reason to pursue this here....! Thank you very
> much
> > for{*filter*} in and being patient....

> > I have EXACTLY what you suggest. These images I was referring to are
> camera
> > shots of what is known as a "Stage Micrometer". Just in case, a stage
> > micrometer is a precisely ruled reference (traceable to US national
Bureau
> > of Standard) necessary to calibrate equipment used to make measurements
at
> > the microscope itself. I want to measure camera shots.... eye fatigue
you
> > know!

> > In this case, the "rulings" on the stage micrometer are in 0.0005"
> > increments. I apologize for US units, I promise to take pictures of the
mm
> > scaled Stage Micrometer!!!!

> > If I understand you correctly:

> > 1) The monitor resolution is not involved. I am only concerned with the
> > magnification of the actual image and
> >     the number of pixels it represents on the screen.

> > 2) At 100X, one graduation mark of 0.0005" will "appear" to be the same
> size
> > as  an 0.05" image taken at 1:1.

> > 3) If the distance for this 0.0005" graduation mark in the image is 100
> > pixels (by the Mouse movement), then the pixel size is
> >     0.0005" / 100 pixels  or  0.00005" per pixel.

> > Now for my extrapolations:

> > 1) The pixel size above is ONLY good for measuring other images taken at
> > 100X, I would have to perform similar calibration for
> >     other magnifications?

> > 2) I can only measure items in images that have known magnifications
(and
> > physical calibrations)? I may not always have this "luxury". Are there
> > techniques for measuring "unknowns"?

> > 3) As long as the complete item to be measured is visible in the image
> (and
> > on the screen) the screen size (1024 by 768 and others) and the image
size
> > (1600 by 1200 and others) are not involved.

> > TIA and I owe you!

> > Gerry



> > > I did actually understand that you are trying to extract the physical
> > > dimensions of an object in your image.

> > > I think the easiest and safest way is to put an object of known
> dimensions
> > > in the microscope, photograph it and then calculate the pixel size
from
> > > that.

> > > You don't need any information about the monitor resolution. The only
> > thing
> > > that matters is the number of pixels. Some screens use 72dpi, others
use
> > > 96dpi, and I don't think those numbers are precise (e.g. a 15in
monitor
> at
> > > VGA resolution will have a different dpi than a 17in monitor at the
same
> > > resolution).

> > > e.g. if a 1mm feature is magnified 100X, it will "appear" to be 100mm.
> I'm
> > > assuming that the camera has some optics and the microscope is not
> > > projecting directly onto the CCD. Assuming the camera can capture the
> full
> > > 100mm image, and for argument's sake let's assume that it covers 1000
of
> > the
> > > 1600 pixels, that means a 1mm feature maps to 1000 pixels, or 1 pixel
> > > represents 1 micrometre.

> > > Does that make sense?

> > > Colin



> > > > Colin

> > > > When I say "measure", I mean I am looking at an image on my monitor
> and
> > > > using the Mouse events to capture X-Y positions which give me the
> > distance
> > > > the Mouse traveled in units of Pixels. I am not physically measuring
> > with
> > > a
> > > > "ruler" or anything like that....I should have mentioned.

> > > > From the Mouse distance in pixels I think I can divide by the screen
> > > > resolution to get a physical size but I'm not sure.....

> > > > Gerry



> > > > > It's always physical resolution. That's what the camera captured.
> I'll
> > > > have
> > > > > to think on this one overnight to figure out how to back out
> physical
> > > > > dimensions.

> > > > > Colin



> > > > > > I'm sorry Colin

> > > > > > The images are taken using a high resolution CCD camera attached
> to
> > a
> > > > Lab
> > > > > > microscope. Although the images are magnified by the
microsocope,
> I
> > > know
> > > > > the
> > > > > > magnification that is used. Using the camera software I can save
> the
> > > > > images
> > > > > > to any size or resolution I desire....well as much as is allowed
> by
> > > the
> > > > > > image format I choose. For example I chose to save them in a
TIFF
> > > format
> > > > > at
> > > > > > 1600 pixels by 1200 pixels at 24 bits per pixel depth. I chose
> this
> > > > format
> > > > > > because I wanted to capture as much detail as possible. The
files
> > are
> > > > > about
> > > > > > 6 MB! I did not want to bore people with those details but if
> > they're
> > > > > > necessary for a good answer...no problem.

> > > > > > I'm not sure how to determine the resolution of these image
files.
> I
> > > was
> > > > > > hoping to use C# to do this for me (DpiX & Y properties) but I'm
> > still
> > > > > > wrestling with "how to do this in an MDI Child form" in other
> posts
> > on
> > > > > this
> > > > > > forum under my name. Regardless, when I view these images in my
> > > program,
> > > > I
> > > > > > believe that I am looking at them at my monitor's resolution
which
> I
> > > > have
> > > > > > set to 120 DPI.

> > > > > > Ignoring the real/apparent magnification effects of capturing
the
> > > images
> > > > > > themselves, when I measure something in an image that is 100
> pixels
> > in
> > > > > > length is it 100 pixels divided by the screen resolution, or 100
> > > pixels
> > > > > > divided by the image's resolution to get at the physical length?

> > > > > > TIA



> > > > > > > The physical dimensions of a image depend on what you are
> planning
> > > on
> > > > > > doing
> > > > > > > with it. The pixel dimensions are absolute. Some image formats
> > carry
> > > > dpi
> > > > > > > information aroung with them, bit you are more or less free to
> do
> > > > > whatever
> > > > > > > you want with the pixels.

> > > > > > > If, for example, the images are aerial photographs, you should
> > have
> > > > the
> > > > > > size
> > > > > > > of the pixel available (e.g. 6m, 10m, etc.) which you can use
to
> > > > convert
> > > > > > to
> > > > > > > physical distances. If you scanned the images, hopefully you
> > > recorded
> > > > > the
> > > > > > > scan settings, and you would use those to get the physical
size.
> > In
> > > > you
> > > > > > are
> > > > > > > actually looking for the physical dimensions of the image on
the
> > > > monitor
> > > > > > you
> > > > > > > need the dpi of the monitor.

> > > > > > > If you provide specifics of exactly what you are doing it
would
> be
> > > > > easier
> > > > > > to
> > > > > > > give you a definite answer.

> > > > > > > Colin



> > > > > > > > I'm writing a small MDI app to measure the size of features
in
> > > TIFF
> > > > > (or
> > > > > > > > others) images that are loaded in pictureboxes on child
forms.
> > > > > Currently
> > > > > > > the
> > > > > > > > code captures these distances in pixels so to convert to a
> > > physical
> > > > > > > > dimension some form of resolution (DpiX/DpiY) information
will
> > be
> > > > > > > necessary.

> > > > > > > > The question in my mind right now is, which resolution
> applies?
> > > The
> > > > > > screen
> > > > > > > > resolution, or the actual image resolution? Both can be
> readily
> > > > > obtained
> > > > > > > in
> > > > > > > > code.

> > > > > > > > My guess is that no matter whether the image resolution is
> > higher
> > > or
> > > > > > lower
> > > > > > > > than the screen resolution, it is the screen resolution that
> > > applies
> > > > > in
> > > > > > > this
> > > > > > > > case.

...

read more »



Wed, 04 May 2005 07:48:14 GMT  
 
 [ 9 post ] 

 Relevant Pages 

1. Printing versus screen resolution....

2. resource size versus screeen resolution

3. image list only allows 16x16 resolution?

4. interpolation alogrithm for increasing resolution of image

5. Screen resolution

6. Dynamic Screen Resolution

7. get screen resolution

8. Handling Screen Resolution Changes

9. Change screen resolution on the fly

10. Change screen resolution to 800x600 at runtime

11. Most widely used screen resolution?

12. help!! with dos screen resolution

 

 
Powered by phpBB® Forum Software