an odd image ID problem
Locked
an odd image ID problem
 
|
Re: an odd image ID problem
|
|
Re: an odd image ID problem
|
|
Re: an odd image ID problem
|
|
Slider: Change Insets
|
|
Re: an odd image ID problem
|
|
Re: an odd image ID problem
|
|
Hi Giovanni,
now I could reproduce it with ImageJ under Java 1.6 on an old (and slow) Mac (MacOS 10.10.5), see the screenshot attached. It does not help to add up to three seconds waiting time after "Stack to Images", and even much later the Windows menu shows the 'old' background file being the current one (screenshot). ImageJ 1.51p and the latest daily build give the same result. I really wonder why... Michael ________________________________________________________________ On 06/10/2017 13:04, Cardone, Giovanni wrote: > Hi Michael, > > my tests were on Windows, but I repeated them on a fresh install of ImageJ on Linux, and the issue remains. > Regarding delays, I don't observe any, the macro runs in less than 200 ms. > I also tested variants of the script and below is the simplest one, showing a consistent result: after 'stack to images' the active image is the last generate before the converted one. This would point to a race conditions like you suspected. > > Best, > Giovanni > > - script > > run("Close All"); > print("\\Clear"); > > newImage("dummy", "32-bit ramp", 64, 32, 3); > dummyID=getImageID(); > currentTitle=getTitle(); > print("First image: Title: ",currentTitle," ID: ",dummyID); > newImage("OriginalStack", "32-bit ramp", 64, 32, 3); > origStackID=getImageID(); > print("origStackID ",origStackID); > print(" - converting stack to images"); > run("Stack to Images"); > afterStacktoIm3ID=getImageID();// current image > currentTitle=getTitle(); > print("active image after stack-to-images: Title: ",currentTitle," ID: ",afterStacktoIm3ID); > > - output from script > > First image: Title: dummy ID: -550 > origStackID -551 > - converting stack to images > active image after stack-to-images: Title: dummy ID: -550 > > > > -----Original Message----- > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Michael Schmid > Sent: Freitag, 6. Oktober 2017 12:04 > To: [hidden email] > Subject: Re: an odd image ID problem > > Hi Giovanni, Jeremy, > > hummmm, I can't reproduce the problem here. I have ImageJ 1.51r15; Java > 1.8.0_112 [64-bit]; Linux 4.4.0-96-generic (Ubuntu 16.04). > Macro results are below. To me, they look fine. > > - Which ImageJ version and operating systems do you use? > - Did you experience an unusual delay running the macro (did it take more than 1.5 seconds for the 'Stack to Images'?) > > I guess that there is a race condition. In Java, images are operated asynchronously; in my experience, the 'windowActivated' of a windowListener can come before the layout of the window is done and the window actually appears on the screen. The ImagePlus.show() method includes code that should work around such problems, but maybe it is not always effective. > > -- > > Nevertheless, I have also experienced rare cases where the image shown as foreground image on the screen was not the same one as that considered foreground image by ImageJ This has happened during interactive work (no macros) and was not reproducible. I have the impression that it might be related to closing a window. (It was not with the current ImageJ version - as I said, it's very rare, and I'm not sure whether it also happens under Linux - I use many platforms). > > > Michael > ________________________________________________________________ > > > The original macro reports: > origStackID -26 > stack-images > after stack-images: current Title: OriginalStack-0003 ID -29 IDs orig -26 after stack-images -29 -28 -27 > > repeat but include image duplication NeworigStackID -30 ID of duplicate -31 > stack-images > after stack-images: current Title: NewOriginalStack-1-0003 > currentID -34 > IDs orig -26 after stack-images -34 -33 -32 ________________________________________________________________ > > The revised macro reports: > origStackID -36 > - converting stack to images > active image after first stack-to-images: Title: OriginalStack-0003 > ID: -39 > > NeworigStackID -40 > - converting new stack to images > active image after second stack-to-images: Title: NewOriginalStack-0004 > currentID -44 > > ________________________________________________________________ > > > On 06/10/2017 10:00, Cardone, Giovanni wrote: >> Hi Jeremy, >> >> I could reproduce the issue, and I pinpointed the problem to the "Stack to Images" command. >> I simplified your example, to show that after the second call to "stack to Images" the active Image is the last one generated by the first call, which to me is an unwanted behavior. >> Hopefully this will help to find the problem in the code. >> >> Best, >> Giovanni >> >> >> run("Close All"); >> print("\\Clear"); >> >> newImage("OriginalStack", "32-bit ramp", 64, 32, 3); >> origStackID=getImageID(); print("origStackID ",origStackID); print(" - >> conveting stack to images"); run("Stack to Images"); >> afterStacktoIm3ID=getImageID();// current image >> currentTitle=getTitle(); print("active image after first >> stack-to-images: Title: ",currentTitle," ID: ",afterStacktoIm3ID); >> >> print(" "); >> newImage("NewOriginalStack", "32-bit ramp", 64, 32, 4); >> NeworigStackID=getImageID(); print("NeworigStackID ",NeworigStackID); >> print(" - conveting new stack to images"); run("Stack to Images"); >> afterSecondStacktoIm3ID=getImageID();// current image >> currentTitle=getTitle(); print("active image after second >> stack-to-images: Title: ",currentTitle," currentID >> ",afterSecondStacktoIm3ID); >> >> >> -----Original Message----- >> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of >> Jeremy Adler >> Sent: Donnerstag, 5. Oktober 2017 10:06 >> To: [hidden email] >> Subject: an odd image ID problem >> >> >> A small problem with the IDs of images or maybe my assumption that the last created image is the current image. >> >> The macro code creates a stack of 3 images then splits the stack. >> The IDs of the 3 daughter images are obtained by finding the ID of the top image (3rd and last image in the stack). >> This is as it should be. >> >> But (second block of code) when I duplicate the stack before splitting the stack, the duplicate stack gets the expected ID. >> But after splitting the stack the IDs of the daughter images are reported incorrectly. >> The problem seems to be that although the final 3rd image in the stack is still the top image, when I check the current ID and title are those of the original stack. >> >> What am I missing ? >> >> run("Close All"); >> print("\\Clear"); >> >> newImage("OriginalStack", "32-bit ramp", 64, 32, 3); origStackID=getImageID();print("origStackID ",origStackID); run("Stack to Images"); >> print(" stack-images"); >> afterStacktoIm3ID=getImageID();// current image >> currentTitle=getTitle();print("after stack-images: current Title: >> ",currentTitle," ID ",afterStacktoIm3ID); >> after2ID=afterStacktoIm3ID+1;// 2nd image from stack >> after1ID=afterStacktoIm3ID+2;// 2nd image from stack/ print("IDs orig >> ",origStackID," after stack-images >> ",afterStacktoIm3ID,after2ID,after1ID); >> selectImage(after2ID); >> >> print(" "); >> print(" repeat but include image duplication"); >> newImage("NewOriginalStack", "32-bit ramp", 64, 32, 3); NeworigStackID=getImageID();print("NeworigStackID ",NeworigStackID); run("Duplicate...", "duplicate"); DupImageID=getImageID(); print("ID of duplicate ",DupImageID); run("Stack to Images"); >> print(" stack-images"); >> afterStacktoIm3ID=getImageID();// current image >> currentTitle=getTitle();print("after stack-images: current Title: >> ",currentTitle," currentID ",afterStacktoIm3ID); >> after2ID=afterStacktoIm3ID+1;// 2nd image from stack >> after1ID=afterStacktoIm3ID+2;// 2nd image from stack/ print("IDs orig >> ",origStackID," after stack-images >> ",afterStacktoIm3ID,after2ID,after1ID); >> selectImage(after2ID); >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
tmp-testImageID.jpg (71K)
Re: an odd image ID problem
|
|
| Free forum by Nabble | Edit this page |