Calibration of image -> wrong results

[w00d]

[ctrueden]

Hi Marius,

Unfortunately, it looks like no text came through with your message.

Everyone: this mail resulted from a chat conversion on IRC:

http://code.imagej.net/chatlogs/fiji-devel?times=prefix&start-date=2014-12-19&start-time=0400&end-date=2014-12-19&end-time=1210

I encouraged Marius to write to the list for guidance.

Hopefully the log above gives enough context to the problem.

Regards,
Curtis

2014-12-19 12:03 GMT-06:00 Marius Krause <[hidden email]>:
>
>

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

[w00d]

@ctrueden thx and here is the text:
Hello,
 
the attached image shows wrong calibration results, is imageJ broken or is my process false?
 
The attached image was calibrated for water to be 1000 kg/m³ (white area)
and air being 0 kg/m³ (black area). But instead water is calculated to be 2000 and air 1000.
What do I have to do to calibrate the image correctly?
 

Initial Expected Results
-992 	0 	2025
2 	1000 	1025

Kind regards Marc

[Gabriel Landini]

On Saturday 20 Dec 2014 17:26:06 you wrote:
> @ctrueden thx and here is the text:
> Hello,
>
> the attached image shows wrong calibration results, is imageJ broken or is
> my process false?

Since water is about 0 before and should be 1000 after
and air is about -1000 before and should be 0 after, then you only need to add
1000 to the greyscale value to obtain the calibrated values.

So in the formula y=a+bx, a=1003 and b (the coefficient) is ~1 and effectively
you are adding 1003 to all pixels.

I have no idea how you obtained the measurements after the calibration but the
calibration plot is correct. Why you get different numbers I do not know.
Are you sure you did not process something twice?
Cheers

Gabriel

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

[Rasband, Wayne (NIH/NIMH) [E]]

On Dec 20, 2014, at 8:26 PM, w00d <[hidden email]> wrote:

>
> @ctrueden thx and here is the text:
> Hello,
>
> the attached image shows wrong calibration results, is imageJ broken or is
> my process false?
>
> The attached image was calibrated for water to be 1000 kg/m³ (white area)
> and air being 0 kg/m³ (black area). But instead water is calculated to be
> 2000 and air 1000.
> What do I have to do to calibrate the image correctly?

You need to remove the existing calibration before measuring “water before calibration” and “air before calibration”. You can do this by setting the “Function” to “None” in the Analyze>Calibrate dialog box. This step is necessary because signed 16-bit images are always calibrated. As you move the cursor over a signed 16-bit image (e.g., the “CT” sample image), you will see the signed (calibrated) value and the uncelebrated value in parenthesis.

-wayne

> Kind regards Marc
>
> <http://imagej.1557.x6.nabble.com/file/n5011015/prob_imagej.png>
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/Calibration-of-image-wrong-results-tp5011006p5011015.html
> Sent from the ImageJ mailing list archive at Nabble.com.



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

[w00d]

@Wayne thank you this was a very helpful tip.

Summary of solution to measure and calibrate density, CT, X-ray images:

1. Delete existing calibration: Analyze-->Calibrate-->select none-->save
2. Measure values: Ctr+M
3. Calibrate: Analyze-->Calibrate-->select straight line-->add unit-->save