Min/max constrast values are lost when duplicating virtual stacks

[Stein Rørvik]

Hello,
In February last year I submitted a bug report about min/max settings in virtual stacks which was fixed by Wayne, see the below post.

I now found another case of what appears to be the same bug:
If you duplicate a virtual stack, the min/max settings are lost if you duplicate the entire stack instead of just a subrange. Here is a macro that reproduces the problem:

//open sample data, and save as separate images in a folder in order to make a virtual stack
run("T1 Head (2.4M, 16-bits)");
File.makeDirectory("C:\\Users\\steinr\\temp\\T1-virtual-stack\\");
run("Image Sequence... ", "format=TIFF save=C:\\Users\\steinr\\temp\\T1-virtual-stack\\");
//close();

run("Image Sequence...", "open=C:\\Users\\steinr\\temp\\T1-virtual-stack\\ sort use");
setSlice(64);
setMinAndMax(1, 300); //the brightness was increased because we say we are interested in the darker tissue

//extract a substack of a sub-range using Duplicate
selectWindow("T1-virtual-stack");
run("Duplicate...", "duplicate range=60-70");
rename("Duplicate range=60-70");
setSlice(nSlices/2);
//the min/max setting was kept

//extract a substack of the full sub-range using Duplicate
selectWindow("T1-virtual-stack");
run("Duplicate...", "duplicate range=1-129");
rename("Duplicate range=1-129");
setSlice(nSlices/2);
//the min/max setting was lost

//extract a substack of a sub-range using Make Substack
selectWindow("T1-virtual-stack");
run("Make Substack...", "slices=60-70");
rename("Substack slices=60-70");
setSlice(nSlices/2);
//the min/max setting was kept

//extract a substack of the full sub-range using Make Substack
selectWindow("T1-virtual-stack");
run("Make Substack...", "slices=1-129");
rename("Substack slices=1-129");
setSlice(nSlices/2);
//the min/max setting was kept

run("Tile");

Stein

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Wayne Rasband
Sent: 16. februar 2017 20:32
To: [hidden email]
Subject: Re: Min/max constrast values are lost when resampling virtual stacks

On Feb 16, 2017, at 7:33 AM, Stein Rørvik <[hidden email]> wrote:
>
> This is a related issue to that described in my last post:
>
> When working with virtual stacks, any min/max constrast values that are set after opening the stack, are lost when rescaling or making a substack.

Thanks to the very helpful test macro, this bug is fixed in the latest ImageJ daily build (1.51k10).

-wayne

> It does not occur with real stacks. rescaling or making a substack. This problem occurs both via then GUI and via macros. It does not occur with real stacks. I assume this is because when working with virtual stacks, if the min/max setting was not already saved in the original files, it is forgotten when the files are re-read from files during the rescale or substack operations.
>
> The reason this (minor) issue is an annoyance for me, is that in my case the virtual stack import is done by a macro that automatically sets the proper min/max values based on material data available in a metafile, so I need to relocate and re-set the proper min/max values whenever I want to scale or make a substack from a virtual stack when using the GUI. In a macro this is not a problem of course, as I then just need to use the getMinAndMax command.
>
> Anyway, the following macro reproduces the problem:
>
> //open sample data, and save as separate images in a folder in order
> to make a virtual stack
> run("T1 Head (2.4M, 16-bits)");
> run("Image Sequence... ", "format=TIFF
> save=C:\\Users\\steinr\\T1-virtual-stack\\");
> close();
>
> run("Image Sequence...", "open=C:\\Users\\steinr\\T1-virtual-stack\\
> sort use"); setSlice(64);
> setMinAndMax(1, 300); //the brightness was increased because we say we are interested in the darker tissue
>
> //extract a substack
> selectWindow("T1-virtual-stack");
> run("Make Substack...", "  slices=60-70");
> setSlice(5); //same as original slice 64
> //the min/max setting was lost
>
> //resample the stack
> selectWindow("T1-virtual-stack");
> run("Scale...", "x=.5 y=.5 z=.5 width=128 height=128 depth=64
> interpolation=Bilinear average process create
> title=T1-virtual-stack-16bit-scaled");
> setSlice(32);
> //the min/max setting was lost
>
>
> //For comparison: When working with true stacks, the min/max constrast values are kept properly:
>
> run("T1 Head (2.4M, 16-bits)");
> setSlice(64);
> setMinAndMax(1, 300);
>
> //extract a substack
> selectWindow("t1-head.tif");
> run("Make Substack...", "  slices=60-70");
> setSlice(5); //same as original slice 64
> //the min/max setting was kept
>
> //resample the stack
> selectWindow("t1-head.tif");
> run("Scale...", "x=.5 y=.5 z=.5 width=128 height=128 depth=64
> interpolation=Bilinear average process create
> title=T1-virtual-stack-16bit-scaled");
> setSlice(32);
> //the min/max setting was kept
>
> run("Tile");
>
>
>
> Stein
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

... [show rest of quote]

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

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

[Wayne Rasband-2]

> On Jan 8, 2018, at 5:35 AM, Stein Rørvik <[hidden email]> wrote:
>
> Hello,
> In February last year I submitted a bug report about min/max settings in virtual stacks which was fixed by Wayne, see the below post.
>
> I now found another case of what appears to be the same bug:
> If you duplicate a virtual stack, the min/max settings are lost if you duplicate the entire stack instead of just a subrange. Here is a macro that reproduces the problem:

This bug is fixed in the latest ImageJ daily build (1.51u6).

-wayne

> //open sample data, and save as separate images in a folder in order to make a virtual stack
> run("T1 Head (2.4M, 16-bits)");
> File.makeDirectory("C:\\Users\\steinr\\temp\\T1-virtual-stack\\");
> run("Image Sequence... ", "format=TIFF save=C:\\Users\\steinr\\temp\\T1-virtual-stack\\");
> //close();
>
> run("Image Sequence...", "open=C:\\Users\\steinr\\temp\\T1-virtual-stack\\ sort use");
> setSlice(64);
> setMinAndMax(1, 300); //the brightness was increased because we say we are interested in the darker tissue
>
> //extract a substack of a sub-range using Duplicate
> selectWindow("T1-virtual-stack");
> run("Duplicate...", "duplicate range=60-70");
> rename("Duplicate range=60-70");
> setSlice(nSlices/2);
> //the min/max setting was kept
>
> //extract a substack of the full sub-range using Duplicate
> selectWindow("T1-virtual-stack");
> run("Duplicate...", "duplicate range=1-129");
> rename("Duplicate range=1-129");
> setSlice(nSlices/2);
> //the min/max setting was lost
>
> //extract a substack of a sub-range using Make Substack
> selectWindow("T1-virtual-stack");
> run("Make Substack...", "slices=60-70");
> rename("Substack slices=60-70");
> setSlice(nSlices/2);
> //the min/max setting was kept
>
> //extract a substack of the full sub-range using Make Substack
> selectWindow("T1-virtual-stack");
> run("Make Substack...", "slices=1-129");
> rename("Substack slices=1-129");
> setSlice(nSlices/2);
> //the min/max setting was kept
>
> run("Tile");
>
> Stein
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Wayne Rasband
> Sent: 16. februar 2017 20:32
> To: [hidden email]
> Subject: Re: Min/max constrast values are lost when resampling virtual stacks
>
> On Feb 16, 2017, at 7:33 AM, Stein Rørvik <[hidden email]> wrote:
>>
>> This is a related issue to that described in my last post:
>>
>> When working with virtual stacks, any min/max constrast values that are set after opening the stack, are lost when rescaling or making a substack.
>
> Thanks to the very helpful test macro, this bug is fixed in the latest ImageJ daily build (1.51k10).
>
> -wayne
>
>
>> It does not occur with real stacks. rescaling or making a substack. This problem occurs both via then GUI and via macros. It does not occur with real stacks. I assume this is because when working with virtual stacks, if the min/max setting was not already saved in the original files, it is forgotten when the files are re-read from files during the rescale or substack operations.
>>
>> The reason this (minor) issue is an annoyance for me, is that in my case the virtual stack import is done by a macro that automatically sets the proper min/max values based on material data available in a metafile, so I need to relocate and re-set the proper min/max values whenever I want to scale or make a substack from a virtual stack when using the GUI. In a macro this is not a problem of course, as I then just need to use the getMinAndMax command.
>>
>> Anyway, the following macro reproduces the problem:
>>
>> //open sample data, and save as separate images in a folder in order
>> to make a virtual stack
>> run("T1 Head (2.4M, 16-bits)");
>> run("Image Sequence... ", "format=TIFF
>> save=C:\\Users\\steinr\\T1-virtual-stack\\");
>> close();
>>
>> run("Image Sequence...", "open=C:\\Users\\steinr\\T1-virtual-stack\\
>> sort use"); setSlice(64);
>> setMinAndMax(1, 300); //the brightness was increased because we say we are interested in the darker tissue
>>
>> //extract a substack
>> selectWindow("T1-virtual-stack");
>> run("Make Substack...", "  slices=60-70");
>> setSlice(5); //same as original slice 64
>> //the min/max setting was lost
>>
>> //resample the stack
>> selectWindow("T1-virtual-stack");
>> run("Scale...", "x=.5 y=.5 z=.5 width=128 height=128 depth=64
>> interpolation=Bilinear average process create
>> title=T1-virtual-stack-16bit-scaled");
>> setSlice(32);
>> //the min/max setting was lost
>>
>>
>> //For comparison: When working with true stacks, the min/max constrast values are kept properly:
>>
>> run("T1 Head (2.4M, 16-bits)");
>> setSlice(64);
>> setMinAndMax(1, 300);
>>
>> //extract a substack
>> selectWindow("t1-head.tif");
>> run("Make Substack...", "  slices=60-70");
>> setSlice(5); //same as original slice 64
>> //the min/max setting was kept
>>
>> //resample the stack
>> selectWindow("t1-head.tif");
>> run("Scale...", "x=.5 y=.5 z=.5 width=128 height=128 depth=64
>> interpolation=Bilinear average process create
>> title=T1-virtual-stack-16bit-scaled");
>> setSlice(32);
>> //the min/max setting was kept
>>
>> run("Tile");
>>
>>
>>
>> Stein

... [show rest of quote]

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