Non-square pixels test

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Non-square pixels test

Gabriel Landini
Hi,
Can somebody point me to what is the proper way to find out in a plugin if the
image is made of square pixels?

Is this testing if cal.pixelWidth() is the same as cal.pixelHeight()?

Any examples?

Many thanks,

Gabriel
Reply | Threaded
Open this post in threaded view
|

AW: Non-square pixels test

Gerhard Holst
Dear Gabriel,

even if any function would exist, allowing for measuring this, what would it mean?
In case of Interline CCDs or CMOS image sensors with microlenses on the pixels, you might find out
with special test setups if the acceptance angle varies between horizontal and vertical.
Nevertheless it is impossible to exactly find out just by the image.

I know that photogrammetry uses calibrated set-ups which are imaged and further
with a lot of calculations and the knowledge about the real dimensions of the imaged object
the image sensor can be precisely calculated.

Best approach maybe the sensor data sheet.

I am in doubt, if this information can be gained just from an image.

With best regards,
Gerhard


_______________________________
 
Dr. Gerhard Holst
Research & Development
PCO AG
Donaupark 11
93309 Kelheim, Germany
fon +49 (0)9441 2005 36
fax +49 (0)9441 2005 20
mob +49 (0)179 9100 319
[hidden email] (off)
[hidden email] (priv)
www.pco.de
 

-----Ursprüngliche Nachricht-----
Von: ImageJ Interest Group [mailto:[hidden email]] Im Auftrag von Gabriel Landini
Gesendet: Freitag, 2. Dezember 2005 13:06
An: [hidden email]
Betreff: Non-square pixels test

Hi,
Can somebody point me to what is the proper way to find out in a plugin if the image is made of square pixels?

Is this testing if cal.pixelWidth() is the same as cal.pixelHeight()?

Any examples?

Many thanks,

Gabriel
Reply | Threaded
Open this post in threaded view
|

Re: AW: Non-square pixels test

Gabriel Landini
On Friday 02 December 2005 13:16, Gerhard Holst wrote:
> even if any function would exist, allowing for measuring this, what would
> it mean?

I would allow to apply a scaling factor to any pixel measurements one can do.

> In case of Interline CCDs or CMOS image sensors with microlenses
> on the pixels, you might find out with special test setups if the
> acceptance angle varies between horizontal and vertical. Nevertheless it is
> impossible to exactly find out just by the image.

Perhaps I put my question wrong. I was talking about the scaling that people
apply to compensate for non-squared pixels. I agree that pixel aspect ratio
cannot be known from an arbitrary image.

> I am in doubt, if this information can be gained just from an image.

It can if the user has added the pixel width and heigh (and this gets stored
in the tiff tags, I believe. The problem is that I do not know what to test
for it.

On Friday 02 December 2005 14:07, Wayne wrote:
> Pixels are nominally square if cal.pixelWidth = cal.pixelHeight.

Great, that is what I wanted.
Many thanks to both.

Gabriel
Reply | Threaded
Open this post in threaded view
|

Re: Non-square pixels test

Prodanov, D. (FYS)
In reply to this post by Gabriel Landini
Hi Gabriel,

I don't see the idea behind such a test. If it tests the camera,
then you can image a square object and read the measurements directly.
Apart from this setup I don't see how you can test if
a pixel is "square" or not. Do you have a particular application in mind?

best regards

Dimiter
Reply | Threaded
Open this post in threaded view
|

Re: Non-square pixels test

Gabriel Landini
On Saturday 03 December 2005 23:26, Prodanov, D. (FYS) wrote:
> I don't see the idea behind such a test. If it tests the camera,
> then you can image a square object and read the measurements directly.
> Apart from this setup I don't see how you can test if
> a pixel is "square" or not. Do you have a particular application in mind?

Hi Dimiter,
Of course you are correct. The problem is that I posed the question
incorrectly.  :-/
I wanted to put a test in the Particles8_Pluse pluing to warn the user about
using the plugin when the image has non-square pixel and he/she has taken
care of calibrating it by inserting the pixel widht and height with the
Image>Properties dialog.

Since my plugin assumes square pixels and it does not use the calibrated
units, using images with aspect ratio pixels different to 1 will return
incorrect results. (the plugin assumes that 2 side-neigbours are 1 unit apart
and 2 corner neighbours sqrt(2) apart).

Until I modify the plugin to use calibrated units (one of these years), this
test is the least I could do to avoid getting flawed measurements.
Wayne's suggestion works fine.

If the images were captured with non-square pixels devices and the user has
not calibrated the image either, then there is not much I can do to prevent
misuse of the plugin.

Thanks for the feedback.
Cheers,

Gabriel
Reply | Threaded
Open this post in threaded view
|

surface pit measurements

Justin Murdock-2
Is there a plugin that can be used to calculate the average width to
depth ratio of pits/cracks on a rock,tile,glass, etc. surface from
confocal microscopy stacks without measuring each individual pit by
hand?

Thanks in advance.

Justin



Justin Murdock
Kansas State University
Division of Biology
232 Ackert Hall
Manhattan, KS 66506