Hey all,
I'd like to give imagej (or a plugin therein) a volume and an ideal radius and have it spit out where the 3D local maxima are. The task is similar to what "Process->Filters->Maximum..." already does in 2D, except I'm only interested in the maxima locations (so single points or a point list would be ideal). I figure there must be a plugin that does exactly this, but I'm at a loss for finding it. Anyone have any ideas? Thanks! Brad |
Hi,
You can have a look to these plugins : http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start Best, Thomas Brad Busse a écrit : > Hey all, > > I'd like to give imagej (or a plugin therein) a volume and an ideal > radius and have it spit out where the 3D local maxima are. The task is > similar to what "Process->Filters->Maximum..." already does in 2D, > except I'm only interested in the maxima locations (so single points or > a point list would be ideal). I figure there must be a plugin that does > exactly this, but I'm at a loss for finding it. Anyone have any ideas? > > Thanks! > Brad > > -- /**********************************************************/ Thomas Boudier, MCU Université Pierre et Marie Curie, IFR 83. Bat B 7ème étage, porte 709, Jussieu. Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 /*******************************************************/ |
All,
These functions seem to be real slow. You might want to try Fiji's Plugins/Process/Maximum (3D), if a 3x3 local neighborhood for defining the local max is OK. Also, this seems to only filter an image, so does not return a list of points. You may need to write a macro or a plugin to do this if something doesn't exist already. David Webster On Mon, Sep 7, 2009 at 3:20 AM, Thomas Boudier < [hidden email]> wrote: > Hi, > > You can have a look to these plugins : > > http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start > > Best, > > Thomas > > > Brad Busse a écrit : > >> Hey all, >> >> I'd like to give imagej (or a plugin therein) a volume and an ideal radius >> and have it spit out where the 3D local maxima are. The task is similar to >> what "Process->Filters->Maximum..." already does in 2D, except I'm only >> interested in the maxima locations (so single points or a point list would >> be ideal). I figure there must be a plugin that does exactly this, but I'm >> at a loss for finding it. Anyone have any ideas? >> >> Thanks! >> Brad >> >> >> > -- > /**********************************************************/ > Thomas Boudier, MCU Université Pierre et Marie Curie, > IFR 83. Bat B 7ème étage, porte 709, Jussieu. > Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 > /*******************************************************/ > |
Hi,
I know this plugin is not very fast, but you can use the maximum filter in my fast JNI filter and compute the difference between the original and max filtered image, a max local should then be at zero in this difference image. Here a small macro that uses 3D object counter to detect the max local pixels and display their coordinates : rename("stack"); // this may have changed fro mthe version o nthe internet ;-) run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); imageCalculator("Difference create stack", "stack","3D Maximum"); run("Invert", "stack"); run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 font_size=10 store_results_within_a_table_named_after_the_image_(macro_friendly) redirect_to=stack"); run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 exclude_objects_on_edges statistics"); The 3D filters in fiji are nice and fast, but it looks like they work on a 3x3x3 cube, I personally prefer to work with spheres. best, Thomas David Webster a écrit : > All, > > These functions seem to be real slow. You might want to try Fiji's > Plugins/Process/Maximum (3D), if a 3x3 local neighborhood for defining the > local max is OK. > Also, this seems to only filter an image, so does not return a list of > points. You may need to write a macro or a plugin to do this if something > doesn't exist already. > > David Webster > On Mon, Sep 7, 2009 at 3:20 AM, Thomas Boudier < > [hidden email]> wrote: > >> Hi, >> >> You can have a look to these plugins : >> >> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start >> >> Best, >> >> Thomas >> >> >> Brad Busse a écrit : >> >>> Hey all, >>> >>> I'd like to give imagej (or a plugin therein) a volume and an ideal radius >>> and have it spit out where the 3D local maxima are. The task is similar to >>> what "Process->Filters->Maximum..." already does in 2D, except I'm only >>> interested in the maxima locations (so single points or a point list would >>> be ideal). I figure there must be a plugin that does exactly this, but I'm >>> at a loss for finding it. Anyone have any ideas? >>> >>> Thanks! >>> Brad >>> >>> >>> >> -- >> /**********************************************************/ >> Thomas Boudier, MCU Université Pierre et Marie Curie, >> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >> /*******************************************************/ >> > > -- /**********************************************************/ Thomas Boudier, MCU Université Pierre et Marie Curie, IFR 83. Bat B 7ème étage, porte 709, Jussieu. Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 /*******************************************************/ |
Thomas,
I tried jni3d_.class. It is much faster. I ran your script with a few mods as there were some name differences. Here it is xr=5; yr=5; zr=5; rename("stack"); // this may have changed fro mthe version o nthe internet ;-) //run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); // Note space after jni3d run("jni3d ", "filter=Maximum radius_x=xr radius_y=yr radius_z=zr"); imageCalculator("Difference create stack", "stack","3D Maximum"); run("Invert", "stack"); //run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 font_size=10 store_results_within_a_table_named_after_the_image_(macro_friendly) redirect_to=stack"); run("Set Measurements", " centroid mean_gray_value dots_size=16 font_size=10 store_results_within_a_table_named_after_the_image_(macro_friendly) redirect_to=stack"); //run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 exclude_objects_on_edges statistics"); run("3D object counter...", "threshold=255 slice=13 min.=1 max.=1134972 exclude_objects_on_edges statistics"); David Webster On Thu, Sep 10, 2009 at 7:14 AM, Thomas Boudier < [hidden email]> wrote: > Hi, > > I know this plugin is not very fast, but you can use the maximum filter in > my fast JNI filter and compute the difference between the original and max > filtered image, a max local should then be at zero in this difference image. > Here a small macro that uses 3D object counter to detect the max local > pixels and display their coordinates : > > rename("stack"); > // this may have changed fro mthe version o nthe internet ;-) > run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); > imageCalculator("Difference create stack", "stack","3D Maximum"); > run("Invert", "stack"); > run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 > font_size=10 > store_results_within_a_table_named_after_the_image_(macro_friendly) > redirect_to=stack"); > run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 > exclude_objects_on_edges statistics"); > > The 3D filters in fiji are nice and fast, but it looks like they work on a > 3x3x3 cube, I personally prefer to work with spheres. > > best, > > > Thomas > > > > David Webster a écrit : > > All, >> >> These functions seem to be real slow. You might want to try Fiji's >> Plugins/Process/Maximum (3D), if a 3x3 local neighborhood for defining the >> local max is OK. >> Also, this seems to only filter an image, so does not return a list of >> points. You may need to write a macro or a plugin to do this if something >> doesn't exist already. >> >> David Webster >> On Mon, Sep 7, 2009 at 3:20 AM, Thomas Boudier < >> [hidden email]> wrote: >> >> Hi, >>> >>> You can have a look to these plugins : >>> >>> >>> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start >>> >>> Best, >>> >>> Thomas >>> >>> >>> Brad Busse a écrit : >>> >>> Hey all, >>>> >>>> I'd like to give imagej (or a plugin therein) a volume and an ideal >>>> radius >>>> and have it spit out where the 3D local maxima are. The task is similar >>>> to >>>> what "Process->Filters->Maximum..." already does in 2D, except I'm only >>>> interested in the maxima locations (so single points or a point list >>>> would >>>> be ideal). I figure there must be a plugin that does exactly this, but >>>> I'm >>>> at a loss for finding it. Anyone have any ideas? >>>> >>>> Thanks! >>>> Brad >>>> >>>> >>>> >>>> -- >>> /**********************************************************/ >>> Thomas Boudier, MCU Université Pierre et Marie Curie, >>> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >>> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >>> /*******************************************************/ >>> >>> >> >> > -- > /**********************************************************/ > Thomas Boudier, MCU Université Pierre et Marie Curie, > IFR 83. Bat B 7ème étage, porte 709, Jussieu. > Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 > /*******************************************************/ > |
Hi,
I think you need to extract the radii values from the " " : run("jni3d ", "filter=Maximum radius_x="+xr+" radius_y="+yr +" radius_z="+zr); Thomas David Webster a écrit : > Thomas, > > I tried jni3d_.class. It is much faster. I ran your script with a few mods > as there were some name differences. Here it is > > > xr=5; yr=5; zr=5; > rename("stack"); > // this may have changed fro mthe version o nthe internet ;-) > //run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); > // Note space after jni3d > run("jni3d ", "filter=Maximum radius_x=xr radius_y=yr radius_z=zr"); > imageCalculator("Difference create stack", "stack","3D Maximum"); > run("Invert", "stack"); > //run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 > font_size=10 > store_results_within_a_table_named_after_the_image_(macro_friendly) > redirect_to=stack"); > run("Set Measurements", " centroid mean_gray_value dots_size=16 > font_size=10 > store_results_within_a_table_named_after_the_image_(macro_friendly) > redirect_to=stack"); > //run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 > exclude_objects_on_edges statistics"); > run("3D object counter...", "threshold=255 slice=13 min.=1 max.=1134972 > exclude_objects_on_edges statistics"); > David Webster > > On Thu, Sep 10, 2009 at 7:14 AM, Thomas Boudier < > [hidden email]> wrote: > >> Hi, >> >> I know this plugin is not very fast, but you can use the maximum filter in >> my fast JNI filter and compute the difference between the original and max >> filtered image, a max local should then be at zero in this difference image. >> Here a small macro that uses 3D object counter to detect the max local >> pixels and display their coordinates : >> >> rename("stack"); >> // this may have changed fro mthe version o nthe internet ;-) >> run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); >> imageCalculator("Difference create stack", "stack","3D Maximum"); >> run("Invert", "stack"); >> run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 >> font_size=10 >> store_results_within_a_table_named_after_the_image_(macro_friendly) >> redirect_to=stack"); >> run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 >> exclude_objects_on_edges statistics"); >> >> The 3D filters in fiji are nice and fast, but it looks like they work on a >> 3x3x3 cube, I personally prefer to work with spheres. >> >> best, >> >> >> Thomas >> >> >> >> David Webster a écrit : >> >> All, >>> These functions seem to be real slow. You might want to try Fiji's >>> Plugins/Process/Maximum (3D), if a 3x3 local neighborhood for defining the >>> local max is OK. >>> Also, this seems to only filter an image, so does not return a list of >>> points. You may need to write a macro or a plugin to do this if something >>> doesn't exist already. >>> >>> David Webster >>> On Mon, Sep 7, 2009 at 3:20 AM, Thomas Boudier < >>> [hidden email]> wrote: >>> >>> Hi, >>>> You can have a look to these plugins : >>>> >>>> >>>> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start >>>> >>>> Best, >>>> >>>> Thomas >>>> >>>> >>>> Brad Busse a écrit : >>>> >>>> Hey all, >>>>> I'd like to give imagej (or a plugin therein) a volume and an ideal >>>>> radius >>>>> and have it spit out where the 3D local maxima are. The task is similar >>>>> to >>>>> what "Process->Filters->Maximum..." already does in 2D, except I'm only >>>>> interested in the maxima locations (so single points or a point list >>>>> would >>>>> be ideal). I figure there must be a plugin that does exactly this, but >>>>> I'm >>>>> at a loss for finding it. Anyone have any ideas? >>>>> >>>>> Thanks! >>>>> Brad >>>>> >>>>> >>>>> >>>>> -- >>>> /**********************************************************/ >>>> Thomas Boudier, MCU Université Pierre et Marie Curie, >>>> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >>>> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >>>> /*******************************************************/ >>>> >>>> >>> >> -- >> /**********************************************************/ >> Thomas Boudier, MCU Université Pierre et Marie Curie, >> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >> /*******************************************************/ >> > > -- /**********************************************************/ Thomas Boudier, MCU Université Pierre et Marie Curie, IFR 83. Bat B 7ème étage, porte 709, Jussieu. Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 /*******************************************************/ |
Thomas,
Hmm! More typos.The xr etc defs were only there so I coulkd time things vs radius more easily. Otherwise, the point of the script to document some changes that needed to be made due to naming differences. In any, Brad solved his another way, so these scripts aren't needed. David On Fri, Sep 11, 2009 at 4:49 AM, Thomas Boudier < [hidden email]> wrote: > Hi, > > I think you need to extract the radii values from the " " : > > run("jni3d ", "filter=Maximum radius_x="+xr+" radius_y="+yr > +" radius_z="+zr); > > Thomas > > > David Webster a écrit : > >> Thomas, >> >> I tried jni3d_.class. It is much faster. I ran your script with a few mods >> as there were some name differences. Here it is >> >> >> xr=5; yr=5; zr=5; >> rename("stack"); >> // this may have changed fro mthe version o nthe internet ;-) >> //run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); >> // Note space after jni3d >> run("jni3d ", "filter=Maximum radius_x=xr radius_y=yr radius_z=zr"); >> imageCalculator("Difference create stack", "stack","3D Maximum"); >> run("Invert", "stack"); >> //run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 >> font_size=10 >> store_results_within_a_table_named_after_the_image_(macro_friendly) >> redirect_to=stack"); >> run("Set Measurements", " centroid mean_gray_value dots_size=16 >> font_size=10 >> store_results_within_a_table_named_after_the_image_(macro_friendly) >> redirect_to=stack"); >> //run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 >> exclude_objects_on_edges statistics"); >> run("3D object counter...", "threshold=255 slice=13 min.=1 max.=1134972 >> exclude_objects_on_edges statistics"); >> David Webster >> >> On Thu, Sep 10, 2009 at 7:14 AM, Thomas Boudier < >> [hidden email]> wrote: >> >> Hi, >>> >>> I know this plugin is not very fast, but you can use the maximum filter >>> in >>> my fast JNI filter and compute the difference between the original and >>> max >>> filtered image, a max local should then be at zero in this difference >>> image. >>> Here a small macro that uses 3D object counter to detect the max local >>> pixels and display their coordinates : >>> >>> rename("stack"); >>> // this may have changed fro mthe version o nthe internet ;-) >>> run("3D Filters", "filter=Maximum radius_x=5 radius_y=5 radius_z=5"); >>> imageCalculator("Difference create stack", "stack","3D Maximum"); >>> run("Invert", "stack"); >>> run("Set 3D Measurements", " centroid mean_gray_value dots_size=16 >>> font_size=10 >>> store_results_within_a_table_named_after_the_image_(macro_friendly) >>> redirect_to=stack"); >>> run("3D Objects Counter...", "threshold=255 slice=13 min.=1 max.=1134972 >>> exclude_objects_on_edges statistics"); >>> >>> The 3D filters in fiji are nice and fast, but it looks like they work on >>> a >>> 3x3x3 cube, I personally prefer to work with spheres. >>> >>> best, >>> >>> >>> Thomas >>> >>> >>> >>> David Webster a écrit : >>> >>> All, >>> >>>> These functions seem to be real slow. You might want to try Fiji's >>>> Plugins/Process/Maximum (3D), if a 3x3 local neighborhood for defining >>>> the >>>> local max is OK. >>>> Also, this seems to only filter an image, so does not return a list of >>>> points. You may need to write a macro or a plugin to do this if >>>> something >>>> doesn't exist already. >>>> >>>> David Webster >>>> On Mon, Sep 7, 2009 at 3:20 AM, Thomas Boudier < >>>> [hidden email]> wrote: >>>> >>>> Hi, >>>> >>>>> You can have a look to these plugins : >>>>> >>>>> >>>>> >>>>> http://imagejdocu.tudor.lu/doku.php?id=plugin:morphology:3d_binary_morphological_filters:start >>>>> >>>>> Best, >>>>> >>>>> Thomas >>>>> >>>>> >>>>> Brad Busse a écrit : >>>>> >>>>> Hey all, >>>>> >>>>>> I'd like to give imagej (or a plugin therein) a volume and an ideal >>>>>> radius >>>>>> and have it spit out where the 3D local maxima are. The task is >>>>>> similar >>>>>> to >>>>>> what "Process->Filters->Maximum..." already does in 2D, except I'm >>>>>> only >>>>>> interested in the maxima locations (so single points or a point list >>>>>> would >>>>>> be ideal). I figure there must be a plugin that does exactly this, >>>>>> but >>>>>> I'm >>>>>> at a loss for finding it. Anyone have any ideas? >>>>>> >>>>>> Thanks! >>>>>> Brad >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> >>>>> /**********************************************************/ >>>>> Thomas Boudier, MCU Université Pierre et Marie Curie, >>>>> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >>>>> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >>>>> /*******************************************************/ >>>>> >>>>> >>>>> >>>> -- >>> /**********************************************************/ >>> Thomas Boudier, MCU Université Pierre et Marie Curie, >>> IFR 83. Bat B 7ème étage, porte 709, Jussieu. >>> Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 >>> /*******************************************************/ >>> >>> >> >> > -- > /**********************************************************/ > Thomas Boudier, MCU Université Pierre et Marie Curie, > IFR 83. Bat B 7ème étage, porte 709, Jussieu. > Tel : 01 44 27 20 11 Fax : 01 44 27 22 91 > /*******************************************************/ > |
Free forum by Nabble | Edit this page |