Quantcast

set scale in ImageJ?

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

set scale in ImageJ?

effect10
Dear ImageJ users,

I am quite a newbie on this front so I hope you can help me.

I have made an image of 2.5x magnification. The program I used for this (research assistent) gives as a calibration:
KalibratiePathologie.doc

I have no idea what is the best way to set this in the set scale function of ImageJ.

Sorry for this newbie question, but I hope someone will be so kind to help me out?
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: set scale in ImageJ?

John Oreopoulos
Hi,

If I'm reading this correctly, it looks like this document is saying  
that at 2.5x magnification, a pixel length of 462 corresponds to a  
real length of 2000 micrometers. I'm not sure what the "marker  
length" represents however. If I'm correct, then in ImageJ you can go  
to the Analyze->Set Scale option and fill in these numbers:

Distance in Pixels = 462
Known Distance = 2000
Pixel Aspect Ratio = 1.0 (usually)
Unit of Length = um

Resulting in a scale of 0.231 pixels per micrometer.

The best way to figure this out on your microscope is to take a  
picture at 2.5x magnification of a object of known dimensions (the  
ruled spacings on a ruler for example or something more precise like  
a hemocytometer grid), draw a line across this distance to get the  
pixel length, and then fill in the appropriate numbers again in the  
Set Scale option. That way you're not depending on some vague  
calibration file like this one to get your image scales right, and  
that way you can confirm if the calibration file is correct as well.

John Oreopoulos


On 11-Dec-08, at 12:31 PM, effect10 wrote:

> Dear ImageJ users,
>
> I am quite a newbie on this front so I hope you can help me.
>
> I have made an image of 2.5x magnification. The program I used for  
> this
> (research assistent) gives as a calibration:
> http://n2.nabble.com/file/n1644260/KalibratiePathologie.doc
> KalibratiePathologie.doc
>
> I have no idea what is the best way to set this in the set scale  
> function of
> ImageJ.
>
> Sorry for this newbie question, but I hope someone will be so kind  
> to help
> me out?
> --
> View this message in context: http://n2.nabble.com/set-scale-in- 
> ImageJ--tp1644260p1644260.html
> Sent from the ImageJ mailing list archive at Nabble.com.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: set scale in ImageJ?

Gabriel Landini
On Thursday 11 December 2008 23:09:04 John Oreopoulos wrote:

> If I'm reading this correctly, it looks like this document is saying
> that at 2.5x magnification, a pixel length of 462 corresponds to a
> real length of 2000 micrometers. I'm not sure what the "marker
> length" represents however. If I'm correct, then in ImageJ you can go
> to the Analyze->Set Scale option and fill in these numbers:
>
> Distance in Pixels = 462
> Known Distance = 2000
> Pixel Aspect Ratio = 1.0 (usually)
> Unit of Length = um
>
> Resulting in a scale of 0.231 pixels per micrometer.

Is this optic microscopy? I think that the above may not be correct. At 2.5
times magnification I doubt that you will get a pixel size of 0.231
micrometres. My optical microscope 2x objective resolution is quoted to be
only 4.19 micrometres and the pixel size I get with one of the cameras is 3.5
(so there is a bit of empty magnification)

What about:
2.5x   2000/462=4.33
5x   2000/926=2.16
20x   500/928=0.54
40x   250/932=0.27
10x  1000/930=1.08

I have no idea what the Marker Length is, but if it is the same for 20x and
10x I doubt that it has to do with the calibration. Perhaps some marker
already in the image.?

Cheers
G.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: set scale in ImageJ?

Justin Walker-3
It will depend completely on the size of the CCD chip in your camera,
the distance from the objective to the CCD, and various other factors.  
The only reliable way of determining scale is to image an object of
known dimensions.

- Justin Walker
University of Maryland

Gabriel Landini wrote:

> On Thursday 11 December 2008 23:09:04 John Oreopoulos wrote:
>  
>> If I'm reading this correctly, it looks like this document is saying
>> that at 2.5x magnification, a pixel length of 462 corresponds to a
>> real length of 2000 micrometers. I'm not sure what the "marker
>> length" represents however. If I'm correct, then in ImageJ you can go
>> to the Analyze->Set Scale option and fill in these numbers:
>>
>> Distance in Pixels = 462
>> Known Distance = 2000
>> Pixel Aspect Ratio = 1.0 (usually)
>> Unit of Length = um
>>
>> Resulting in a scale of 0.231 pixels per micrometer.
>>    
>
> Is this optic microscopy? I think that the above may not be correct. At 2.5
> times magnification I doubt that you will get a pixel size of 0.231
> micrometres. My optical microscope 2x objective resolution is quoted to be
> only 4.19 micrometres and the pixel size I get with one of the cameras is 3.5
> (so there is a bit of empty magnification)
>
> What about:
> 2.5x   2000/462=4.33
> 5x   2000/926=2.16
> 20x   500/928=0.54
> 40x   250/932=0.27
> 10x  1000/930=1.08
>
> I have no idea what the Marker Length is, but if it is the same for 20x and
> 10x I doubt that it has to do with the calibration. Perhaps some marker
> already in the image.?
>
> Cheers
> G.
>  
Loading...