Scale is SOMETIMES lost during import

[jumpfunky]

Dear List,


I've encountered a problem when I try to import a set of images.

To reproduce please download the following set of images:

https://dl.dropboxusercontent.com/u/560426/TEST.zip (it's currently
uploading)

Now apply File -> Import -> Image Sequence and check "Use Virtual
Stack". The data is correctly scaled to 0.66 x 0.66 µm.

Now apply File -> Import -> Image Sequence again but do not check "Use
Virtual Stack". Now the scale is lost.

If you delete L'45AA7.tif the data will be correctly imported with the
second approach as well.

Did I found a bug?

Another question:
If you open a single image you will the that the unit is given in cm.
But if you import it with the first approach, the unit is given in µm.
Does ImageJ internally convert the units?

Best,
Thorsten

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

[Herbie]

Dear Thorsten,

I can confirm your findings but is it really necessary to use such a huge stack for test purposes?

"unit is given in cm"

Yes, but this is a matter of taste. The values are correct anyway.

Best

Herbie

:::::::::::::::::::::::::::::::::::::::::::::
Am 10.10.16 um 12:20 schrieb Thorsten Wagner:
> Dear List,
>
>
> I've encountered a problem when I try to import a set of images.
>
> To reproduce please download the following set of images:
>
> https://dl.dropboxusercontent.com/u/560426/TEST.zip (it's currently
> uploading)
>
> Now apply File -> Import -> Image Sequence and check "Use Virtual
> Stack". The data is correctly scaled to 0.66 x 0.66 µm.
>
> Now apply File -> Import -> Image Sequence again but do not check "Use
> Virtual Stack". Now the scale is lost.
>
> If you delete L'45AA7.tif the data will be correctly imported with the
> second approach as well.
>
> Did I found a bug?
>
> Another question:
> If you open a single image you will the that the unit is given in cm.
> But if you import it with the first approach, the unit is given in µm.
> Does ImageJ internally convert the units?
>
> Best,
> Thorsten

[jumpfunky]

Dear Herbert,

thanks for reproducing!

Sorry for that big stack. This smaller stack will work as well:
https://dl.dropboxusercontent.com/u/560426/TEST2.zip

As for the units:

I know that the values are correct. But I was wondering if this
behaviour could be configured like "convert any unit to µm"?

Best,
Thorsten 

> Dear Thorsten,
>
> I can confirm your findings but is it really necessary to use such a
> huge
> stack for test purposes?
>
> "unit is given in cm"
>
> Yes, but this is a matter of taste. The values are correct anyway.
>
> Best
>
> Herbie
>
> :::::::::::::::::::::::::::::::::::::::::::::
> Am 10.10.16 um 12:20 schrieb Thorsten Wagner:
> >
> > Dear List,
> >
> >
> > I've encountered a problem when I try to import a set of images.
> >
> > To reproduce please download the following set of images:
> >
> > https://dl.dropboxusercontent.com/u/560426/TEST.zip (it's currently
> > uploading)
> >
> > Now apply File -> Import -> Image Sequence and check "Use Virtual
> > Stack". The data is correctly scaled to 0.66 x 0.66 µm.
> >
> > Now apply File -> Import -> Image Sequence again but do not check
> > "Use
> > Virtual Stack". Now the scale is lost.
> >
> > If you delete L'45AA7.tif the data will be correctly imported with
> > the
> > second approach as well.
> >
> > Did I found a bug?
> >
> > Another question:
> > If you open a single image you will the that the unit is given in
> > cm.
> > But if you import it with the first approach, the unit is given in
> > µm.
> > Does ImageJ internally convert the units?
> >
> > Best,
> > Thorsten
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/Scale-
> is-SOMETIMES-lost-during-import-tp5017331p5017332.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

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

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

> On Oct 10, 2016, at 6:20 AM, Thorsten Wagner <[hidden email]> wrote:
>
> Dear List,
>
> I've encountered a problem when I try to import a set of images.
>
> To reproduce please download the following set of images:
>
> https://dl.dropboxusercontent.com/u/560426/TEST.zip (it's currently
> uploading)
>
> Now apply File -> Import -> Image Sequence and check "Use Virtual
> Stack". The data is correctly scaled to 0.66 x 0.66 µm.
>
> Now apply File -> Import -> Image Sequence again but do not check "Use
> Virtual Stack". Now the scale is lost.
>
> If you delete L'45AA7.tif the data will be correctly imported with the
> second approach as well.
>
> Did I found a bug?

This bug is fixed in the latest ImageJ daily build (1.51h12).

> Another question:
> If you open a single image you will the that the unit is given in cm.
> But if you import it with the first approach, the unit is given in µm.
> Does ImageJ internally convert the units?

Units are converted from cm to µm when the pixel width is less than 0.0001 cm. Earlier versions of ImageJ only did this conversion when importing a sequence of images as a virtual stack.

-wayne


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