Disjoint ROI selection using macro language
Locked
 
|
Hi All,
I think I have managed to completely confuse myself and would love some help getting straightened out... (1) As I manually select an item in the ROI Manager list it is not the item highlighted in the mask. For example, highlighting the first item in the ROI Manager list will outline the second particle in the mask. (I often confuse the indices between roiManager("select", x); and getResults("Something", x); where x is either zero-based or one-based. This is going to really drive me bonkers.) (2) I expected the total area of the new combined ROI to equal the sum of the areas in the selection (they do not equal - in fact, it looks like only the first particle is added.) A year ago I thought I had this all figured out, but now I can't get it to behave as I would like. I have written a macro (below) is to select some arbitrary number of ROIs and combine them into one. //START MACRO run("Blobs (25K)"); setAutoThreshold(); //run("Threshold..."); setThreshold(125, 255); run("Convert to Mask"); run("Set Measurements...", "area mean standard modal min centroid center perimeter bounding fit shape feret's integrated median skewness kurtosis redirect=None decimal=3"); run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 show=Nothing display clear record add"); //combine items 0, 3 and 7 into one ROI index = newArray(0,3,7); area = 0; for (i=0; i<index.length; i++){ a = getResult("Area", index[i]); area = area + a; print("area[", index[i], "]=", a); } print("Expected total area for 3 items=", area); setKeyDown("shift") roiManager("select", 0); setKeyDown("shift") roiManager("select", 3); setKeyDown("shift") roiManager("select", 7); roiManager("Add"); roiManager("Measure"); print("after measure, nResults=", nResults ," area=" + getResult("Area", nResults-1)); //END MACRO I note there is a bug-fix in ROI manager regarding combining ROIs for v1.42f, but I am running v1.42l. Ben Tupper Java properties applets can read: java.version: 1.5.0_16 java.vendor: Apple Inc. mrj.version: 1050.1.5.0_16-284 os.name: Mac OS X os.version: 10.5.6 os.arch: i386 file.separator: / path.separator: : line.separator: <lf> Java properties only applications can read: user.name: ben user.home: /Users/ben user.dir: /Applications/ImageJ user.country: US file.encoding: MacRoman java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ 1.5.0/Home java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/System/ Library/Java/Extensions/QTJava.zip java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/ Home/lib/ext java.io.tmpdir: /tmp Other properties: IJ.getVersion: 1.42l IJ.isJava2: true IJ.isJava15: true IJ.isJava16: false IJ.isLinux: false IJ.isMacintosh: true IJ.isMacOSX: true IJ.isWindows: false IJ.isVista: false IJ.is64Bit: false Menus.getPlugInsPath: /Applications/ImageJ/plugins/ Menus.getMacrosPath: /Applications/ImageJ/macros/ Prefs.getHomeDir: /Applications/ImageJ Prefs.getThreads: 2 (2 cores) Prefs.open100Percent: false Prefs.blackBackground: false Prefs.useJFileChooser: false Prefs.weightedColor: false Prefs.blackCanvas: false Prefs.pointAutoMeasure: false Prefs.pointAutoNextSlice: false Prefs.requireControlKey: false Prefs.useInvertingLut: false Prefs.antialiasedTools: true Prefs.useInvertingLut: false Prefs.intelByteOrder: false Prefs.doubleBuffer: false Prefs.noPointLabels: false Prefs.disableUndo: false Prefs dir: /Users/ben/Library/Preferences Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ Sample images dir: http://rsb.info.nih.gov/ij/images/ Screen size: 1440x900 Memory in use: 11MB of 400MB (2%) |
Re: Disjoint ROI selection using macro language
|
|
Hi Ben,
using a very similar system (but MacOS 10.4.11, not 10.5), I have neither problems with 1.42l nor with the nightly build of a few days ago. Selecting the ROIs manually works as expected. After running the macro, the log says: area[ 0 ]= 433 area[ 3 ]= 434 area[ 7 ]= 278 Expected total area for 3 items= 1145 after measure, nResults= 129 area=1145 ImageJ Properties: Java properties applets can read: java.version: 1.5.0_16 java.vendor: Apple Computer, Inc. mrj.version: 1040.1.5.0_16-275 os.name: Mac OS X os.version: 10.4.11 os.arch: i386 file.separator: / path.separator: : line.separator: <lf> Java properties only applications can read: (...) user.country: US file.encoding: MacRoman java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ 1.5.0/Home java.class.path: (...) /ij.jar:/System/Library/Java/Extensions/ QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/ Classes/.compatibility/14compatibility.jar java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/ Home/lib/ext java.io.tmpdir: /tmp Other properties: (... same as yours) Prefs.weightedColor: true Prefs.blackCanvas: false Prefs.pointAutoMeasure: false Prefs.pointAutoNextSlice: false Prefs.requireControlKey: false Prefs.useInvertingLut: false Prefs.antialiasedTools: true Prefs.useInvertingLut: false Prefs.intelByteOrder: false Prefs.doubleBuffer: false Prefs.noPointLabels: true Prefs.disableUndo: false (...) Sample images dir: http://rsb.info.nih.gov/ij/images/ Screen size: 1680x1050 Memory in use: 3738K of 800MB (<1%) Is it really 1.42l or a nightly build previous to 1.42l that you are using? The released 1.42l ij.jar has 1.373.997 bytes. Maybe something got stuck - does the problem go away if you restart ImageJ? Michael ________________________________________________________________ On 16 Apr 2009, at 19:44, Ben Tupper wrote: > Hi All, > > I think I have managed to completely confuse myself and would love > some help getting straightened out... > > (1) As I manually select an item in the ROI Manager list it is not > the item highlighted in the mask. For example, highlighting the > first item in the ROI Manager list will outline the second particle > in the mask. (I often confuse the indices between roiManager > ("select", x); and getResults("Something", x); where x is either > zero-based or one-based. This is going to really drive me bonkers.) > > (2) I expected the total area of the new combined ROI to equal the > sum of the areas in the selection (they do not equal - in fact, it > looks like only the first particle is added.) A year ago I thought > I had this all figured out, but now I can't get it to behave as I > would like. I have written a macro (below) is to select some > arbitrary number of ROIs and combine them into one. > > //START MACRO > run("Blobs (25K)"); > setAutoThreshold(); > //run("Threshold..."); > setThreshold(125, 255); > run("Convert to Mask"); > run("Set Measurements...", "area mean standard modal min centroid > center perimeter bounding fit shape feret's integrated median > skewness kurtosis redirect=None decimal=3"); > run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 > show=Nothing display clear record add"); > > //combine items 0, 3 and 7 into one ROI > index = newArray(0,3,7); > area = 0; > for (i=0; i<index.length; i++){ > a = getResult("Area", index[i]); > area = area + a; > print("area[", index[i], "]=", a); > } > print("Expected total area for 3 items=", area); > > setKeyDown("shift") > roiManager("select", 0); > setKeyDown("shift") > roiManager("select", 3); > setKeyDown("shift") > roiManager("select", 7); > > roiManager("Add"); > roiManager("Measure"); > print("after measure, nResults=", nResults ," area=" + getResult > ("Area", nResults-1)); > > //END MACRO > > I note there is a bug-fix in ROI manager regarding combining ROIs > for v1.42f, but I am running v1.42l. > > > Ben Tupper > > > > Java properties applets can read: > java.version: 1.5.0_16 > java.vendor: Apple Inc. > mrj.version: 1050.1.5.0_16-284 > os.name: Mac OS X > os.version: 10.5.6 > os.arch: i386 > file.separator: / > path.separator: : > line.separator: <lf> > > Java properties only applications can read: > user.name: ben > user.home: /Users/ben > user.dir: /Applications/ImageJ > user.country: US > file.encoding: MacRoman > java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ > 1.5.0/Home > java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/ > System/Library/Java/Extensions/QTJava.zip > java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ > Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ > 1.5.0/Home/lib/ext > java.io.tmpdir: /tmp > > Other properties: > IJ.getVersion: 1.42l > IJ.isJava2: true > IJ.isJava15: true > IJ.isJava16: false > IJ.isLinux: false > IJ.isMacintosh: true > IJ.isMacOSX: true > IJ.isWindows: false > IJ.isVista: false > IJ.is64Bit: false > Menus.getPlugInsPath: /Applications/ImageJ/plugins/ > Menus.getMacrosPath: /Applications/ImageJ/macros/ > Prefs.getHomeDir: /Applications/ImageJ > Prefs.getThreads: 2 (2 cores) > Prefs.open100Percent: false > Prefs.blackBackground: false > Prefs.useJFileChooser: false > Prefs.weightedColor: false > Prefs.blackCanvas: false > Prefs.pointAutoMeasure: false > Prefs.pointAutoNextSlice: false > Prefs.requireControlKey: false > Prefs.useInvertingLut: false > Prefs.antialiasedTools: true > Prefs.useInvertingLut: false > Prefs.intelByteOrder: false > Prefs.doubleBuffer: false > Prefs.noPointLabels: false > Prefs.disableUndo: false > Prefs dir: /Users/ben/Library/Preferences > Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ > Sample images dir: http://rsb.info.nih.gov/ij/images/ > Screen size: 1440x900 > Memory in use: 11MB of 400MB (2%) |
|
|
Hi Michael,
Thanks for responding so quickly. The first issue is a result of my own boneheadedness. Note that the left most particle looks like it is labeled "0". It is actually "10" but the label was cutoff and I fooled myself. I can confirm that the ij.jar is the size you specify and that ImageJ thinks it is v1.42l. Restarting ImageJ seems to make no difference. Can you explain how you made those selections manually? I can't make that happen manually or in the macro. Thanks again, Ben On Apr 16, 2009, at 2:23 PM, Michael Schmid wrote: > Hi Ben, > > using a very similar system (but MacOS 10.4.11, not 10.5), I have > neither problems with 1.42l nor with the nightly build of a few days > ago. > > Selecting the ROIs manually works as expected. > > After running the macro, the log says: > area[ 0 ]= 433 > area[ 3 ]= 434 > area[ 7 ]= 278 > Expected total area for 3 items= 1145 > after measure, nResults= 129 area=1145 > > ImageJ Properties: > > Java properties applets can read: > java.version: 1.5.0_16 > java.vendor: Apple Computer, Inc. > mrj.version: 1040.1.5.0_16-275 > os.name: Mac OS X > os.version: 10.4.11 > os.arch: i386 > file.separator: / > path.separator: : > line.separator: <lf> > > Java properties only applications can read: > (...) > user.country: US > file.encoding: MacRoman > java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ > 1.5.0/Home > java.class.path: (...) /ij.jar:/System/Library/Java/Extensions/ > QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/ > 1.5.0/Classes/.compatibility/14compatibility.jar > java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ > Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ > 1.5.0/Home/lib/ext > java.io.tmpdir: /tmp > > Other properties: > (... same as yours) > Prefs.weightedColor: true > Prefs.blackCanvas: false > Prefs.pointAutoMeasure: false > Prefs.pointAutoNextSlice: false > Prefs.requireControlKey: false > Prefs.useInvertingLut: false > Prefs.antialiasedTools: true > Prefs.useInvertingLut: false > Prefs.intelByteOrder: false > Prefs.doubleBuffer: false > Prefs.noPointLabels: true > Prefs.disableUndo: false > (...) > Sample images dir: http://rsb.info.nih.gov/ij/images/ > Screen size: 1680x1050 > Memory in use: 3738K of 800MB (<1%) > > Is it really 1.42l or a nightly build previous to 1.42l that you are > using? > The released 1.42l ij.jar has 1.373.997 bytes. > > Maybe something got stuck - does the problem go away if you restart > ImageJ? > > > Michael > ________________________________________________________________ > > On 16 Apr 2009, at 19:44, Ben Tupper wrote: > >> Hi All, >> >> I think I have managed to completely confuse myself and would love >> some help getting straightened out... >> >> (1) As I manually select an item in the ROI Manager list it is not >> the item highlighted in the mask. For example, highlighting the >> first item in the ROI Manager list will outline the second particle >> in the mask. (I often confuse the indices between >> roiManager("select", x); and getResults("Something", x); where x is >> either zero-based or one-based. This is going to really drive me >> bonkers.) >> >> (2) I expected the total area of the new combined ROI to equal the >> sum of the areas in the selection (they do not equal - in fact, it >> looks like only the first particle is added.) A year ago I thought >> I had this all figured out, but now I can't get it to behave as I >> would like. I have written a macro (below) is to select some >> arbitrary number of ROIs and combine them into one. >> >> //START MACRO >> run("Blobs (25K)"); >> setAutoThreshold(); >> //run("Threshold..."); >> setThreshold(125, 255); >> run("Convert to Mask"); >> run("Set Measurements...", "area mean standard modal min centroid >> center perimeter bounding fit shape feret's integrated median >> skewness kurtosis redirect=None decimal=3"); >> run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 >> show=Nothing display clear record add"); >> >> //combine items 0, 3 and 7 into one ROI >> index = newArray(0,3,7); >> area = 0; >> for (i=0; i<index.length; i++){ >> a = getResult("Area", index[i]); >> area = area + a; >> print("area[", index[i], "]=", a); >> } >> print("Expected total area for 3 items=", area); >> >> setKeyDown("shift") >> roiManager("select", 0); >> setKeyDown("shift") >> roiManager("select", 3); >> setKeyDown("shift") >> roiManager("select", 7); >> >> roiManager("Add"); >> roiManager("Measure"); >> print("after measure, nResults=", nResults ," area=" + >> getResult("Area", nResults-1)); >> >> //END MACRO >> >> I note there is a bug-fix in ROI manager regarding combining ROIs >> for v1.42f, but I am running v1.42l. >> >> >> Ben Tupper >> >> >> >> Java properties applets can read: >> java.version: 1.5.0_16 >> java.vendor: Apple Inc. >> mrj.version: 1050.1.5.0_16-284 >> os.name: Mac OS X >> os.version: 10.5.6 >> os.arch: i386 >> file.separator: / >> path.separator: : >> line.separator: <lf> >> >> Java properties only applications can read: >> user.name: ben >> user.home: /Users/ben >> user.dir: /Applications/ImageJ >> user.country: US >> file.encoding: MacRoman >> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >> 1.5.0/Home >> java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/System/ >> Library/Java/Extensions/QTJava.zip >> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >> 1.5.0/Home/lib/ext >> java.io.tmpdir: /tmp >> >> Other properties: >> IJ.getVersion: 1.42l >> IJ.isJava2: true >> IJ.isJava15: true >> IJ.isJava16: false >> IJ.isLinux: false >> IJ.isMacintosh: true >> IJ.isMacOSX: true >> IJ.isWindows: false >> IJ.isVista: false >> IJ.is64Bit: false >> Menus.getPlugInsPath: /Applications/ImageJ/plugins/ >> Menus.getMacrosPath: /Applications/ImageJ/macros/ >> Prefs.getHomeDir: /Applications/ImageJ >> Prefs.getThreads: 2 (2 cores) >> Prefs.open100Percent: false >> Prefs.blackBackground: false >> Prefs.useJFileChooser: false >> Prefs.weightedColor: false >> Prefs.blackCanvas: false >> Prefs.pointAutoMeasure: false >> Prefs.pointAutoNextSlice: false >> Prefs.requireControlKey: false >> Prefs.useInvertingLut: false >> Prefs.antialiasedTools: true >> Prefs.useInvertingLut: false >> Prefs.intelByteOrder: false >> Prefs.doubleBuffer: false >> Prefs.noPointLabels: false >> Prefs.disableUndo: false >> Prefs dir: /Users/ben/Library/Preferences >> Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ >> Sample images dir: http://rsb.info.nih.gov/ij/images/ >> Screen size: 1440x900 >> Memory in use: 11MB of 400MB (2%) Ben Tupper |
Re: Disjoint ROI selection using macro language
|
|
Hi Ben,
the ROI Manager items have labels 0001-... to 0064-..., corresponding to the labels in the binary image, but they have line numbers 0 to 63 - maybe this causes part of the confusion? I did succeed in selecting three items manually (lines 0, 3, 7, i.e. 0001-..., 0004.. and 0008-...) together with command-click and then use More>Combine. It is a bit nasty since the context menus interferes with it. Measure on the combined ROI gave me an area of 1145 again. I don't know how to do this in a macro; at least I am not aware of a setKeyDown("command") or setKeyDown("ctrl") With "shift", one would select a contiguous range of lines. Best wishes, Michael ________________________________________________________________ On 16 Apr 2009, at 21:03, Ben Tupper wrote: > Hi Michael, > > Thanks for responding so quickly. > > The first issue is a result of my own boneheadedness. Note that > the left most particle looks like it is labeled "0". It is > actually "10" but the label was cutoff and I fooled myself. > > I can confirm that the ij.jar is the size you specify and that > ImageJ thinks it is v1.42l. Restarting ImageJ seems to make no > difference. > > Can you explain how you made those selections manually? I can't > make that happen manually or in the macro. > > Thanks again, > Ben > > > > > > On Apr 16, 2009, at 2:23 PM, Michael Schmid wrote: > >> Hi Ben, >> >> using a very similar system (but MacOS 10.4.11, not 10.5), I have >> neither problems with 1.42l nor with the nightly build of a few >> days ago. >> >> Selecting the ROIs manually works as expected. >> >> After running the macro, the log says: >> area[ 0 ]= 433 >> area[ 3 ]= 434 >> area[ 7 ]= 278 >> Expected total area for 3 items= 1145 >> after measure, nResults= 129 area=1145 >> >> ImageJ Properties: >> >> Java properties applets can read: >> java.version: 1.5.0_16 >> java.vendor: Apple Computer, Inc. >> mrj.version: 1040.1.5.0_16-275 >> os.name: Mac OS X >> os.version: 10.4.11 >> os.arch: i386 >> file.separator: / >> path.separator: : >> line.separator: <lf> >> >> Java properties only applications can read: >> (...) >> user.country: US >> file.encoding: MacRoman >> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >> 1.5.0/Home >> java.class.path: (...) /ij.jar:/System/Library/Java/Extensions/ >> QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/ >> 1.5.0/Classes/.compatibility/14compatibility.jar >> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >> 1.5.0/Home/lib/ext >> java.io.tmpdir: /tmp >> >> Other properties: >> (... same as yours) >> Prefs.weightedColor: true >> Prefs.blackCanvas: false >> Prefs.pointAutoMeasure: false >> Prefs.pointAutoNextSlice: false >> Prefs.requireControlKey: false >> Prefs.useInvertingLut: false >> Prefs.antialiasedTools: true >> Prefs.useInvertingLut: false >> Prefs.intelByteOrder: false >> Prefs.doubleBuffer: false >> Prefs.noPointLabels: true >> Prefs.disableUndo: false >> (...) >> Sample images dir: http://rsb.info.nih.gov/ij/images/ >> Screen size: 1680x1050 >> Memory in use: 3738K of 800MB (<1%) >> >> Is it really 1.42l or a nightly build previous to 1.42l that you >> are using? >> The released 1.42l ij.jar has 1.373.997 bytes. >> >> Maybe something got stuck - does the problem go away if you >> restart ImageJ? >> >> >> Michael >> ________________________________________________________________ >> >> On 16 Apr 2009, at 19:44, Ben Tupper wrote: >> >>> Hi All, >>> >>> I think I have managed to completely confuse myself and would >>> love some help getting straightened out... >>> >>> (1) As I manually select an item in the ROI Manager list it is >>> not the item highlighted in the mask. For example, highlighting >>> the first item in the ROI Manager list will outline the second >>> particle in the mask. (I often confuse the indices between >>> roiManager("select", x); and getResults("Something", x); where x >>> is either zero-based or one-based. This is going to really drive >>> me bonkers.) >>> >>> (2) I expected the total area of the new combined ROI to equal >>> the sum of the areas in the selection (they do not equal - in >>> fact, it looks like only the first particle is added.) A year >>> ago I thought I had this all figured out, but now I can't get it >>> to behave as I would like. I have written a macro (below) is to >>> select some arbitrary number of ROIs and combine them into one. >>> >>> //START MACRO >>> run("Blobs (25K)"); >>> setAutoThreshold(); >>> //run("Threshold..."); >>> setThreshold(125, 255); >>> run("Convert to Mask"); >>> run("Set Measurements...", "area mean standard modal min centroid >>> center perimeter bounding fit shape feret's integrated median >>> skewness kurtosis redirect=None decimal=3"); >>> run("Analyze Particles...", "size=0-Infinity >>> circularity=0.00-1.00 show=Nothing display clear record add"); >>> >>> //combine items 0, 3 and 7 into one ROI >>> index = newArray(0,3,7); >>> area = 0; >>> for (i=0; i<index.length; i++){ >>> a = getResult("Area", index[i]); >>> area = area + a; >>> print("area[", index[i], "]=", a); >>> } >>> print("Expected total area for 3 items=", area); >>> >>> setKeyDown("shift") >>> roiManager("select", 0); >>> setKeyDown("shift") >>> roiManager("select", 3); >>> setKeyDown("shift") >>> roiManager("select", 7); >>> >>> roiManager("Add"); >>> roiManager("Measure"); >>> print("after measure, nResults=", nResults ," area=" + getResult >>> ("Area", nResults-1)); >>> >>> //END MACRO >>> >>> I note there is a bug-fix in ROI manager regarding combining ROIs >>> for v1.42f, but I am running v1.42l. >>> >>> >>> Ben Tupper >>> >>> >>> >>> Java properties applets can read: >>> java.version: 1.5.0_16 >>> java.vendor: Apple Inc. >>> mrj.version: 1050.1.5.0_16-284 >>> os.name: Mac OS X >>> os.version: 10.5.6 >>> os.arch: i386 >>> file.separator: / >>> path.separator: : >>> line.separator: <lf> >>> >>> Java properties only applications can read: >>> user.name: ben >>> user.home: /Users/ben >>> user.dir: /Applications/ImageJ >>> user.country: US >>> file.encoding: MacRoman >>> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >>> 1.5.0/Home >>> java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/ >>> System/Library/Java/Extensions/QTJava.zip >>> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >>> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >>> 1.5.0/Home/lib/ext >>> java.io.tmpdir: /tmp >>> >>> Other properties: >>> IJ.getVersion: 1.42l >>> IJ.isJava2: true >>> IJ.isJava15: true >>> IJ.isJava16: false >>> IJ.isLinux: false >>> IJ.isMacintosh: true >>> IJ.isMacOSX: true >>> IJ.isWindows: false >>> IJ.isVista: false >>> IJ.is64Bit: false >>> Menus.getPlugInsPath: /Applications/ImageJ/plugins/ >>> Menus.getMacrosPath: /Applications/ImageJ/macros/ >>> Prefs.getHomeDir: /Applications/ImageJ >>> Prefs.getThreads: 2 (2 cores) >>> Prefs.open100Percent: false >>> Prefs.blackBackground: false >>> Prefs.useJFileChooser: false >>> Prefs.weightedColor: false >>> Prefs.blackCanvas: false >>> Prefs.pointAutoMeasure: false >>> Prefs.pointAutoNextSlice: false >>> Prefs.requireControlKey: false >>> Prefs.useInvertingLut: false >>> Prefs.antialiasedTools: true >>> Prefs.useInvertingLut: false >>> Prefs.intelByteOrder: false >>> Prefs.doubleBuffer: false >>> Prefs.noPointLabels: false >>> Prefs.disableUndo: false >>> Prefs dir: /Users/ben/Library/Preferences >>> Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ >>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>> Screen size: 1440x900 >>> Memory in use: 11MB of 400MB (2%) > > Ben Tupper |
|
|
On Apr 16, 2009, at 3:22 PM, Michael Schmid wrote:
> Hi Ben, > > the ROI Manager items have labels 0001-... to 0064-..., > corresponding to the labels in the binary image, but they have line > numbers 0 to 63 - maybe this causes part of the confusion? Yes, I am always sawing back-and-forth between different indexing schemes. It is very taxing on my walnut-sized-brain. It wasn't so bad when it was just Java and IDL (0-based) but now throw in R and the results table (1-based)... argh! > > I did succeed in selecting three items manually (lines 0, 3, 7, i.e. > 0001-..., 0004.. and 0008-...) together with command-click and then > use More>Combine. It is a bit nasty since the context menus > interferes with it. Measure on the combined ROI gave me an area of > 1145 again. OK, I can do this now and get the same result as you using the click- command. With the fictional click-command it works as long as the menu command "Measure" is used - not the roiManager("Measure") command. roiManager("Select", 0); with click-command roiManager("Select", 3); with click-command roiManager("Select", 7); with click-command roiManager("Combine"); run("Measure"); <--- note: it isn't roiManager("Measure"); I had been assuming that I could run "Analyze particles..." with additions to the ROI Manager. Then I thought I could step through the items in the list to see if they meet my qualifications - finally I would select only those that qualify. Since the click-command is fictional, I'll have to change my filtering so that I only add those items that qualify. Thanks again for your help! Ben > > I don't know how to do this in a macro; at least I am not aware of a > setKeyDown("command") or setKeyDown("ctrl") > > With "shift", one would select a contiguous range of lines. > > > Best wishes, > > Michael > ________________________________________________________________ > > On 16 Apr 2009, at 21:03, Ben Tupper wrote: > >> Hi Michael, >> >> Thanks for responding so quickly. >> >> The first issue is a result of my own boneheadedness. Note that >> the left most particle looks like it is labeled "0". It is >> actually "10" but the label was cutoff and I fooled myself. >> >> I can confirm that the ij.jar is the size you specify and that >> ImageJ thinks it is v1.42l. Restarting ImageJ seems to make no >> difference. >> >> Can you explain how you made those selections manually? I can't >> make that happen manually or in the macro. >> >> Thanks again, >> Ben >> >> >> >> >> >> On Apr 16, 2009, at 2:23 PM, Michael Schmid wrote: >> >>> Hi Ben, >>> >>> using a very similar system (but MacOS 10.4.11, not 10.5), I have >>> neither problems with 1.42l nor with the nightly build of a few >>> days ago. >>> >>> Selecting the ROIs manually works as expected. >>> >>> After running the macro, the log says: >>> area[ 0 ]= 433 >>> area[ 3 ]= 434 >>> area[ 7 ]= 278 >>> Expected total area for 3 items= 1145 >>> after measure, nResults= 129 area=1145 >>> >>> ImageJ Properties: >>> >>> Java properties applets can read: >>> java.version: 1.5.0_16 >>> java.vendor: Apple Computer, Inc. >>> mrj.version: 1040.1.5.0_16-275 >>> os.name: Mac OS X >>> os.version: 10.4.11 >>> os.arch: i386 >>> file.separator: / >>> path.separator: : >>> line.separator: <lf> >>> >>> Java properties only applications can read: >>> (...) >>> user.country: US >>> file.encoding: MacRoman >>> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >>> 1.5.0/Home >>> java.class.path: (...) /ij.jar:/System/Library/Java/Extensions/ >>> QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/ >>> 1.5.0/Classes/.compatibility/14compatibility.jar >>> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >>> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >>> 1.5.0/Home/lib/ext >>> java.io.tmpdir: /tmp >>> >>> Other properties: >>> (... same as yours) >>> Prefs.weightedColor: true >>> Prefs.blackCanvas: false >>> Prefs.pointAutoMeasure: false >>> Prefs.pointAutoNextSlice: false >>> Prefs.requireControlKey: false >>> Prefs.useInvertingLut: false >>> Prefs.antialiasedTools: true >>> Prefs.useInvertingLut: false >>> Prefs.intelByteOrder: false >>> Prefs.doubleBuffer: false >>> Prefs.noPointLabels: true >>> Prefs.disableUndo: false >>> (...) >>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>> Screen size: 1680x1050 >>> Memory in use: 3738K of 800MB (<1%) >>> >>> Is it really 1.42l or a nightly build previous to 1.42l that you >>> are using? >>> The released 1.42l ij.jar has 1.373.997 bytes. >>> >>> Maybe something got stuck - does the problem go away if you >>> restart ImageJ? >>> >>> >>> Michael >>> ________________________________________________________________ >>> >>> On 16 Apr 2009, at 19:44, Ben Tupper wrote: >>> >>>> Hi All, >>>> >>>> I think I have managed to completely confuse myself and would >>>> love some help getting straightened out... >>>> >>>> (1) As I manually select an item in the ROI Manager list it is >>>> not the item highlighted in the mask. For example, highlighting >>>> the first item in the ROI Manager list will outline the second >>>> particle in the mask. (I often confuse the indices between >>>> roiManager("select", x); and getResults("Something", x); where x >>>> is either zero-based or one-based. This is going to really drive >>>> me bonkers.) >>>> >>>> (2) I expected the total area of the new combined ROI to equal >>>> the sum of the areas in the selection (they do not equal - in >>>> fact, it looks like only the first particle is added.) A year >>>> ago I thought I had this all figured out, but now I can't get it >>>> to behave as I would like. I have written a macro (below) is to >>>> select some arbitrary number of ROIs and combine them into one. >>>> >>>> //START MACRO >>>> run("Blobs (25K)"); >>>> setAutoThreshold(); >>>> //run("Threshold..."); >>>> setThreshold(125, 255); >>>> run("Convert to Mask"); >>>> run("Set Measurements...", "area mean standard modal min centroid >>>> center perimeter bounding fit shape feret's integrated median >>>> skewness kurtosis redirect=None decimal=3"); >>>> run("Analyze Particles...", "size=0-Infinity >>>> circularity=0.00-1.00 show=Nothing display clear record add"); >>>> >>>> //combine items 0, 3 and 7 into one ROI >>>> index = newArray(0,3,7); >>>> area = 0; >>>> for (i=0; i<index.length; i++){ >>>> a = getResult("Area", index[i]); >>>> area = area + a; >>>> print("area[", index[i], "]=", a); >>>> } >>>> print("Expected total area for 3 items=", area); >>>> >>>> setKeyDown("shift") >>>> roiManager("select", 0); >>>> setKeyDown("shift") >>>> roiManager("select", 3); >>>> setKeyDown("shift") >>>> roiManager("select", 7); >>>> >>>> roiManager("Add"); >>>> roiManager("Measure"); >>>> print("after measure, nResults=", nResults ," area=" + >>>> getResult("Area", nResults-1)); >>>> >>>> //END MACRO >>>> >>>> I note there is a bug-fix in ROI manager regarding combining ROIs >>>> for v1.42f, but I am running v1.42l. >>>> >>>> >>>> Ben Tupper >>>> >>>> >>>> >>>> Java properties applets can read: >>>> java.version: 1.5.0_16 >>>> java.vendor: Apple Inc. >>>> mrj.version: 1050.1.5.0_16-284 >>>> os.name: Mac OS X >>>> os.version: 10.5.6 >>>> os.arch: i386 >>>> file.separator: / >>>> path.separator: : >>>> line.separator: <lf> >>>> >>>> Java properties only applications can read: >>>> user.name: ben >>>> user.home: /Users/ben >>>> user.dir: /Applications/ImageJ >>>> user.country: US >>>> file.encoding: MacRoman >>>> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >>>> 1.5.0/Home >>>> java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/ >>>> System/Library/Java/Extensions/QTJava.zip >>>> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >>>> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >>>> 1.5.0/Home/lib/ext >>>> java.io.tmpdir: /tmp >>>> >>>> Other properties: >>>> IJ.getVersion: 1.42l >>>> IJ.isJava2: true >>>> IJ.isJava15: true >>>> IJ.isJava16: false >>>> IJ.isLinux: false >>>> IJ.isMacintosh: true >>>> IJ.isMacOSX: true >>>> IJ.isWindows: false >>>> IJ.isVista: false >>>> IJ.is64Bit: false >>>> Menus.getPlugInsPath: /Applications/ImageJ/plugins/ >>>> Menus.getMacrosPath: /Applications/ImageJ/macros/ >>>> Prefs.getHomeDir: /Applications/ImageJ >>>> Prefs.getThreads: 2 (2 cores) >>>> Prefs.open100Percent: false >>>> Prefs.blackBackground: false >>>> Prefs.useJFileChooser: false >>>> Prefs.weightedColor: false >>>> Prefs.blackCanvas: false >>>> Prefs.pointAutoMeasure: false >>>> Prefs.pointAutoNextSlice: false >>>> Prefs.requireControlKey: false >>>> Prefs.useInvertingLut: false >>>> Prefs.antialiasedTools: true >>>> Prefs.useInvertingLut: false >>>> Prefs.intelByteOrder: false >>>> Prefs.doubleBuffer: false >>>> Prefs.noPointLabels: false >>>> Prefs.disableUndo: false >>>> Prefs dir: /Users/ben/Library/Preferences >>>> Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ >>>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>>> Screen size: 1440x900 >>>> Memory in use: 11MB of 400MB (2%) >> >> Ben Tupper Ben Tupper |
Re: Disjoint ROI selection using macro language
|
|
Ben Tupper a écrit :
> On Apr 16, 2009, at 3:22 PM, Michael Schmid wrote: > >> Hi Ben, >> >> the ROI Manager items have labels 0001-... to 0064-..., corresponding >> to the labels in the binary image, but they have line numbers 0 to 63 >> - maybe this causes part of the confusion? > > Yes, I am always sawing back-and-forth between different indexing > schemes. It is very taxing on my walnut-sized-brain. It wasn't so bad > when it was just Java and IDL (0-based) but now throw in R and the > results table (1-based)... argh! > > >> >> I did succeed in selecting three items manually (lines 0, 3, 7, i.e. >> 0001-..., 0004.. and 0008-...) together with command-click and then >> use More>Combine. It is a bit nasty since the context menus >> interferes with it. Measure on the combined ROI gave me an area of >> 1145 again. > > > OK, I can do this now and get the same result as you using the > click-command. With the fictional click-command it works as long as > the menu command "Measure" is used - not the roiManager("Measure") > command. > > roiManager("Select", 0); with click-command > roiManager("Select", 3); with click-command > roiManager("Select", 7); with click-command > roiManager("Combine"); > run("Measure"); <--- note: it isn't roiManager("Measure"); > > > I had been assuming that I could run "Analyze particles..." with > additions to the ROI Manager. Then I thought I could step through the > items in the list to see if they meet my qualifications - finally I > would select only those that qualify. Since the click-command is > fictional, I'll have to change my filtering so that I only add those > items that qualify. > > Thanks again for your help! > Ben > > > >> >> I don't know how to do this in a macro; at least I am not aware of a >> setKeyDown("command") or setKeyDown("ctrl") >> >> With "shift", one would select a contiguous range of lines. >> >> >> Best wishes, >> >> Michael >> ________________________________________________________________ >> >> On 16 Apr 2009, at 21:03, Ben Tupper wrote: >> >>> Hi Michael, >>> >>> Thanks for responding so quickly. >>> >>> The first issue is a result of my own boneheadedness. Note that >>> the left most particle looks like it is labeled "0". It is actually >>> "10" but the label was cutoff and I fooled myself. >>> >>> I can confirm that the ij.jar is the size you specify and that >>> ImageJ thinks it is v1.42l. Restarting ImageJ seems to make no >>> difference. >>> >>> Can you explain how you made those selections manually? I can't >>> make that happen manually or in the macro. >>> >>> Thanks again, >>> Ben >>> >>> >>> >>> >>> >>> On Apr 16, 2009, at 2:23 PM, Michael Schmid wrote: >>> >>>> Hi Ben, >>>> >>>> using a very similar system (but MacOS 10.4.11, not 10.5), I have >>>> neither problems with 1.42l nor with the nightly build of a few >>>> days ago. >>>> >>>> Selecting the ROIs manually works as expected. >>>> >>>> After running the macro, the log says: >>>> area[ 0 ]= 433 >>>> area[ 3 ]= 434 >>>> area[ 7 ]= 278 >>>> Expected total area for 3 items= 1145 >>>> after measure, nResults= 129 area=1145 >>>> >>>> ImageJ Properties: >>>> >>>> Java properties applets can read: >>>> java.version: 1.5.0_16 >>>> java.vendor: Apple Computer, Inc. >>>> mrj.version: 1040.1.5.0_16-275 >>>> os.name: Mac OS X >>>> os.version: 10.4.11 >>>> os.arch: i386 >>>> file.separator: / >>>> path.separator: : >>>> line.separator: <lf> >>>> >>>> Java properties only applications can read: >>>> (...) >>>> user.country: US >>>> file.encoding: MacRoman >>>> java.home: >>>> /System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Home >>>> java.class.path: (...) >>>> /ij.jar:/System/Library/Java/Extensions/QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Classes/.compatibility/14compatibility.jar >>>> >>>> java.ext.dirs: >>>> /Library/Java/Extensions:/System/Library/Java/Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Home/lib/ext >>>> >>>> java.io.tmpdir: /tmp >>>> >>>> Other properties: >>>> (... same as yours) >>>> Prefs.weightedColor: true >>>> Prefs.blackCanvas: false >>>> Prefs.pointAutoMeasure: false >>>> Prefs.pointAutoNextSlice: false >>>> Prefs.requireControlKey: false >>>> Prefs.useInvertingLut: false >>>> Prefs.antialiasedTools: true >>>> Prefs.useInvertingLut: false >>>> Prefs.intelByteOrder: false >>>> Prefs.doubleBuffer: false >>>> Prefs.noPointLabels: true >>>> Prefs.disableUndo: false >>>> (...) >>>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>>> Screen size: 1680x1050 >>>> Memory in use: 3738K of 800MB (<1%) >>>> >>>> Is it really 1.42l or a nightly build previous to 1.42l that you >>>> are using? >>>> The released 1.42l ij.jar has 1.373.997 bytes. >>>> >>>> Maybe something got stuck - does the problem go away if you restart >>>> ImageJ? >>>> >>>> >>>> Michael >>>> ________________________________________________________________ >>>> >>>> On 16 Apr 2009, at 19:44, Ben Tupper wrote: >>>> >>>>> Hi All, >>>>> >>>>> I think I have managed to completely confuse myself and would love >>>>> some help getting straightened out... >>>>> >>>>> (1) As I manually select an item in the ROI Manager list it is not >>>>> the item highlighted in the mask. For example, highlighting the >>>>> first item in the ROI Manager list will outline the second >>>>> particle in the mask. (I often confuse the indices between >>>>> roiManager("select", x); and getResults("Something", x); where x >>>>> is either zero-based or one-based. This is going to really drive >>>>> me bonkers.) >>>>> >>>>> (2) I expected the total area of the new combined ROI to equal the >>>>> sum of the areas in the selection (they do not equal - in fact, it >>>>> looks like only the first particle is added.) A year ago I >>>>> thought I had this all figured out, but now I can't get it to >>>>> behave as I would like. I have written a macro (below) is to >>>>> select some arbitrary number of ROIs and combine them into one. >>>>> >>>>> //START MACRO >>>>> run("Blobs (25K)"); >>>>> setAutoThreshold(); >>>>> //run("Threshold..."); >>>>> setThreshold(125, 255); >>>>> run("Convert to Mask"); >>>>> run("Set Measurements...", "area mean standard modal min centroid >>>>> center perimeter bounding fit shape feret's integrated median >>>>> skewness kurtosis redirect=None decimal=3"); >>>>> run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 >>>>> show=Nothing display clear record add"); >>>>> >>>>> //combine items 0, 3 and 7 into one ROI >>>>> index = newArray(0,3,7); >>>>> area = 0; >>>>> for (i=0; i<index.length; i++){ >>>>> a = getResult("Area", index[i]); >>>>> area = area + a; >>>>> print("area[", index[i], "]=", a); >>>>> } >>>>> print("Expected total area for 3 items=", area); >>>>> >>>>> setKeyDown("shift") >>>>> roiManager("select", 0); >>>>> setKeyDown("shift") >>>>> roiManager("select", 3); >>>>> setKeyDown("shift") >>>>> roiManager("select", 7); >>>>> >>>>> roiManager("Add"); >>>>> roiManager("Measure"); >>>>> print("after measure, nResults=", nResults ," area=" + >>>>> getResult("Area", nResults-1)); >>>>> >>>>> //END MACRO >>>>> >>>>> I note there is a bug-fix in ROI manager regarding combining ROIs >>>>> for v1.42f, but I am running v1.42l. >>>>> >>>>> >>>>> Ben Tupper >>>>> >>>>> >>>>> >>>>> Java properties applets can read: >>>>> java.version: 1.5.0_16 >>>>> java.vendor: Apple Inc. >>>>> mrj.version: 1050.1.5.0_16-284 >>>>> os.name: Mac OS X >>>>> os.version: 10.5.6 >>>>> os.arch: i386 >>>>> file.separator: / >>>>> path.separator: : >>>>> line.separator: <lf> >>>>> >>>>> Java properties only applications can read: >>>>> user.name: ben >>>>> user.home: /Users/ben >>>>> user.dir: /Applications/ImageJ >>>>> user.country: US >>>>> file.encoding: MacRoman >>>>> java.home: >>>>> /System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Home >>>>> java.class.path: >>>>> /Applications/ImageJ/ImageJ.app/../ij.jar:/System/Library/Java/Extensions/QTJava.zip >>>>> >>>>> java.ext.dirs: >>>>> /Library/Java/Extensions:/System/Library/Java/Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Home/lib/ext >>>>> >>>>> java.io.tmpdir: /tmp >>>>> >>>>> Other properties: >>>>> IJ.getVersion: 1.42l >>>>> IJ.isJava2: true >>>>> IJ.isJava15: true >>>>> IJ.isJava16: false >>>>> IJ.isLinux: false >>>>> IJ.isMacintosh: true >>>>> IJ.isMacOSX: true >>>>> IJ.isWindows: false >>>>> IJ.isVista: false >>>>> IJ.is64Bit: false >>>>> Menus.getPlugInsPath: /Applications/ImageJ/plugins/ >>>>> Menus.getMacrosPath: /Applications/ImageJ/macros/ >>>>> Prefs.getHomeDir: /Applications/ImageJ >>>>> Prefs.getThreads: 2 (2 cores) >>>>> Prefs.open100Percent: false >>>>> Prefs.blackBackground: false >>>>> Prefs.useJFileChooser: false >>>>> Prefs.weightedColor: false >>>>> Prefs.blackCanvas: false >>>>> Prefs.pointAutoMeasure: false >>>>> Prefs.pointAutoNextSlice: false >>>>> Prefs.requireControlKey: false >>>>> Prefs.useInvertingLut: false >>>>> Prefs.antialiasedTools: true >>>>> Prefs.useInvertingLut: false >>>>> Prefs.intelByteOrder: false >>>>> Prefs.doubleBuffer: false >>>>> Prefs.noPointLabels: false >>>>> Prefs.disableUndo: false >>>>> Prefs dir: /Users/ben/Library/Preferences >>>>> Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ >>>>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>>>> Screen size: 1440x900 >>>>> Memory in use: 11MB of 400MB (2%) >>> >>> Ben Tupper > > Ben Tupper i'm sorry to not have the answer to your problem but I take the opportunity to talk about a problem I have. I'm segmenting cells and add them to the manager what I would do next is extract each cell to generate a new image by using the corresponding ROI. Did someone allready tried such an approach ? Thanks, Fabrice. -- Senger Fabrice |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by BenTupper
Hi Ben,
maybe deleting those particles that don't match from the roi manager would be a solution? Michael ________________________________________________________________ On 16 Apr 2009, at 21:52, Ben Tupper wrote: > On Apr 16, 2009, at 3:22 PM, Michael Schmid wrote: > >> Hi Ben, >> >> the ROI Manager items have labels 0001-... to 0064-..., >> corresponding to the labels in the binary image, but they have >> line numbers 0 to 63 - maybe this causes part of the confusion? > > Yes, I am always sawing back-and-forth between different indexing > schemes. It is very taxing on my walnut-sized-brain. It wasn't so > bad when it was just Java and IDL (0-based) but now throw in R and > the results table (1-based)... argh! > > >> >> I did succeed in selecting three items manually (lines 0, 3, 7, >> i.e. 0001-..., 0004.. and 0008-...) together with command-click >> and then use More>Combine. It is a bit nasty since the context >> menus interferes with it. Measure on the combined ROI gave me an >> area of 1145 again. > > > OK, I can do this now and get the same result as you using the > click-command. With the fictional click-command it works as long > as the menu command "Measure" is used - not the roiManager > ("Measure") command. > > roiManager("Select", 0); with click-command > roiManager("Select", 3); with click-command > roiManager("Select", 7); with click-command > roiManager("Combine"); > run("Measure"); <--- note: it isn't roiManager("Measure"); > > > I had been assuming that I could run "Analyze particles..." with > additions to the ROI Manager. Then I thought I could step through > the items in the list to see if they meet my qualifications - > finally I would select only those that qualify. Since the click- > command is fictional, I'll have to change my filtering so that I > only add those items that qualify. > > Thanks again for your help! > Ben > > > >> >> I don't know how to do this in a macro; at least I am not aware of >> a setKeyDown("command") or setKeyDown("ctrl") >> >> With "shift", one would select a contiguous range of lines. >> >> >> Best wishes, >> >> Michael >> ________________________________________________________________ >> >> On 16 Apr 2009, at 21:03, Ben Tupper wrote: >> >>> Hi Michael, >>> >>> Thanks for responding so quickly. >>> >>> The first issue is a result of my own boneheadedness. Note that >>> the left most particle looks like it is labeled "0". It is >>> actually "10" but the label was cutoff and I fooled myself. >>> >>> I can confirm that the ij.jar is the size you specify and that >>> ImageJ thinks it is v1.42l. Restarting ImageJ seems to make no >>> difference. >>> >>> Can you explain how you made those selections manually? I can't >>> make that happen manually or in the macro. >>> >>> Thanks again, >>> Ben >>> >>> >>> >>> >>> >>> On Apr 16, 2009, at 2:23 PM, Michael Schmid wrote: >>> >>>> Hi Ben, >>>> >>>> using a very similar system (but MacOS 10.4.11, not 10.5), I >>>> have neither problems with 1.42l nor with the nightly build of a >>>> few days ago. >>>> >>>> Selecting the ROIs manually works as expected. >>>> >>>> After running the macro, the log says: >>>> area[ 0 ]= 433 >>>> area[ 3 ]= 434 >>>> area[ 7 ]= 278 >>>> Expected total area for 3 items= 1145 >>>> after measure, nResults= 129 area=1145 >>>> >>>> ImageJ Properties: >>>> >>>> Java properties applets can read: >>>> java.version: 1.5.0_16 >>>> java.vendor: Apple Computer, Inc. >>>> mrj.version: 1040.1.5.0_16-275 >>>> os.name: Mac OS X >>>> os.version: 10.4.11 >>>> os.arch: i386 >>>> file.separator: / >>>> path.separator: : >>>> line.separator: <lf> >>>> >>>> Java properties only applications can read: >>>> (...) >>>> user.country: US >>>> file.encoding: MacRoman >>>> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >>>> 1.5.0/Home >>>> java.class.path: (...) /ij.jar:/System/Library/Java/Extensions/ >>>> QTJava.zip:/System/Library/Frameworks/JavaVM.framework/Versions/ >>>> 1.5.0/Classes/.compatibility/14compatibility.jar >>>> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >>>> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >>>> 1.5.0/Home/lib/ext >>>> java.io.tmpdir: /tmp >>>> >>>> Other properties: >>>> (... same as yours) >>>> Prefs.weightedColor: true >>>> Prefs.blackCanvas: false >>>> Prefs.pointAutoMeasure: false >>>> Prefs.pointAutoNextSlice: false >>>> Prefs.requireControlKey: false >>>> Prefs.useInvertingLut: false >>>> Prefs.antialiasedTools: true >>>> Prefs.useInvertingLut: false >>>> Prefs.intelByteOrder: false >>>> Prefs.doubleBuffer: false >>>> Prefs.noPointLabels: true >>>> Prefs.disableUndo: false >>>> (...) >>>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>>> Screen size: 1680x1050 >>>> Memory in use: 3738K of 800MB (<1%) >>>> >>>> Is it really 1.42l or a nightly build previous to 1.42l that you >>>> are using? >>>> The released 1.42l ij.jar has 1.373.997 bytes. >>>> >>>> Maybe something got stuck - does the problem go away if you >>>> restart ImageJ? >>>> >>>> >>>> Michael >>>> ________________________________________________________________ >>>> >>>> On 16 Apr 2009, at 19:44, Ben Tupper wrote: >>>> >>>>> Hi All, >>>>> >>>>> I think I have managed to completely confuse myself and would >>>>> love some help getting straightened out... >>>>> >>>>> (1) As I manually select an item in the ROI Manager list it is >>>>> not the item highlighted in the mask. For example, highlighting >>>>> the first item in the ROI Manager list will outline the second >>>>> particle in the mask. (I often confuse the indices between >>>>> roiManager("select", x); and getResults("Something", x); where >>>>> x is either zero-based or one-based. This is going to really >>>>> drive me bonkers.) >>>>> >>>>> (2) I expected the total area of the new combined ROI to equal >>>>> the sum of the areas in the selection (they do not equal - in >>>>> fact, it looks like only the first particle is added.) A year >>>>> ago I thought I had this all figured out, but now I can't get >>>>> it to behave as I would like. I have written a macro (below) >>>>> is to select some arbitrary number of ROIs and combine them >>>>> into one. >>>>> >>>>> //START MACRO >>>>> run("Blobs (25K)"); >>>>> setAutoThreshold(); >>>>> //run("Threshold..."); >>>>> setThreshold(125, 255); >>>>> run("Convert to Mask"); >>>>> run("Set Measurements...", "area mean standard modal min >>>>> centroid center perimeter bounding fit shape feret's integrated >>>>> median skewness kurtosis redirect=None decimal=3"); >>>>> run("Analyze Particles...", "size=0-Infinity >>>>> circularity=0.00-1.00 show=Nothing display clear record add"); >>>>> >>>>> //combine items 0, 3 and 7 into one ROI >>>>> index = newArray(0,3,7); >>>>> area = 0; >>>>> for (i=0; i<index.length; i++){ >>>>> a = getResult("Area", index[i]); >>>>> area = area + a; >>>>> print("area[", index[i], "]=", a); >>>>> } >>>>> print("Expected total area for 3 items=", area); >>>>> >>>>> setKeyDown("shift") >>>>> roiManager("select", 0); >>>>> setKeyDown("shift") >>>>> roiManager("select", 3); >>>>> setKeyDown("shift") >>>>> roiManager("select", 7); >>>>> >>>>> roiManager("Add"); >>>>> roiManager("Measure"); >>>>> print("after measure, nResults=", nResults ," area=" + getResult >>>>> ("Area", nResults-1)); >>>>> >>>>> //END MACRO >>>>> >>>>> I note there is a bug-fix in ROI manager regarding combining >>>>> ROIs for v1.42f, but I am running v1.42l. >>>>> >>>>> >>>>> Ben Tupper >>>>> >>>>> >>>>> >>>>> Java properties applets can read: >>>>> java.version: 1.5.0_16 >>>>> java.vendor: Apple Inc. >>>>> mrj.version: 1050.1.5.0_16-284 >>>>> os.name: Mac OS X >>>>> os.version: 10.5.6 >>>>> os.arch: i386 >>>>> file.separator: / >>>>> path.separator: : >>>>> line.separator: <lf> >>>>> >>>>> Java properties only applications can read: >>>>> user.name: ben >>>>> user.home: /Users/ben >>>>> user.dir: /Applications/ImageJ >>>>> user.country: US >>>>> file.encoding: MacRoman >>>>> java.home: /System/Library/Frameworks/JavaVM.framework/Versions/ >>>>> 1.5.0/Home >>>>> java.class.path: /Applications/ImageJ/ImageJ.app/../ij.jar:/ >>>>> System/Library/Java/Extensions/QTJava.zip >>>>> java.ext.dirs: /Library/Java/Extensions:/System/Library/Java/ >>>>> Extensions:/System/Library/Frameworks/JavaVM.framework/Versions/ >>>>> 1.5.0/Home/lib/ext >>>>> java.io.tmpdir: /tmp >>>>> >>>>> Other properties: >>>>> IJ.getVersion: 1.42l >>>>> IJ.isJava2: true >>>>> IJ.isJava15: true >>>>> IJ.isJava16: false >>>>> IJ.isLinux: false >>>>> IJ.isMacintosh: true >>>>> IJ.isMacOSX: true >>>>> IJ.isWindows: false >>>>> IJ.isVista: false >>>>> IJ.is64Bit: false >>>>> Menus.getPlugInsPath: /Applications/ImageJ/plugins/ >>>>> Menus.getMacrosPath: /Applications/ImageJ/macros/ >>>>> Prefs.getHomeDir: /Applications/ImageJ >>>>> Prefs.getThreads: 2 (2 cores) >>>>> Prefs.open100Percent: false >>>>> Prefs.blackBackground: false >>>>> Prefs.useJFileChooser: false >>>>> Prefs.weightedColor: false >>>>> Prefs.blackCanvas: false >>>>> Prefs.pointAutoMeasure: false >>>>> Prefs.pointAutoNextSlice: false >>>>> Prefs.requireControlKey: false >>>>> Prefs.useInvertingLut: false >>>>> Prefs.antialiasedTools: true >>>>> Prefs.useInvertingLut: false >>>>> Prefs.intelByteOrder: false >>>>> Prefs.doubleBuffer: false >>>>> Prefs.noPointLabels: false >>>>> Prefs.disableUndo: false >>>>> Prefs dir: /Users/ben/Library/Preferences >>>>> Current dir: /Users/Shared/data/441.2009-04-23.400A4X_03 2/ >>>>> Sample images dir: http://rsb.info.nih.gov/ij/images/ >>>>> Screen size: 1440x900 >>>>> Memory in use: 11MB of 400MB (2%) >>> >>> Ben Tupper > > Ben Tupper |
|
|
In reply to this post by Fabrice Senger
On Apr 17, 2009, at 4:08 AM, Senger Fabrice wrote:
>> > Hi there, > > i'm sorry to not have the answer to your problem but I take the > opportunity to talk about a problem I have. I'm segmenting cells and > add them to the manager what I would do next is extract each cell to > generate a new image by using the corresponding ROI. > Did someone allready tried such an approach ? > > Thanks, > > Fabrice. > > Hi Fabrice, If you want to transfer a region from one image to another then you could do it with something like the following (note that ROI Manager is not used). //START MACRO run("Blobs (25K)"); setAutoThreshold(); //run("Threshold..."); setThreshold(125, 255); run("Convert to Mask"); maskID = getImageID(); run("Set Measurements...", "area mean standard modal min centroid center perimeter bounding fit shape feret's integrated median skewness kurtosis redirect=None decimal=3"); run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 show=Nothing display clear record "); //first method creates a new image to accept the subimage ix = 22; bx = getResult("BX", ix); by = getResult("BY", ix); w = getResult("Width", ix); h = getResult("Height", ix); name = "#" + ix+1; newImage(name, "8-bit White", w,h,1); img23ID = getImageID(); selectImage(maskID); makeRectangle(bx,by,w,h); run("Copy"); selectImage(img23ID); run("Paste"); run("Invert"); // the second method exploits the File > New > Internal Clipboard menu item ix = 30; bx = getResult("BX", ix); by = getResult("BY", ix); w = getResult("Width", ix); h = getResult("Height", ix); selectImage(maskID); makeRectangle(bx,by,w,h); run("Copy"); name = "#" + ix+1; run("Internal Clipboard"); rename(name); //END MACRO Ben Tupper |
|
|
In reply to this post by Michael Schmid
On Apr 17, 2009, at 4:23 AM, Michael Schmid wrote:
> Hi Ben, > > maybe deleting those particles that don't match from the roi manager > would be a solution? That might be a solution - I have not tried that. In the meantime, I think I have figured out a way to do this - but the approach is to add to the Manager only those particles that I want in there. It turns it into a "front-end" problem rather than a "back-end" problem. I was trying the back end solution because I mistakenly thought that I needed to use the ROI manager is I wanted to use... xs = getResult("XStart", i); ys = getResult("YStart", i); doWand(xs, ys); getSelectionCoordinates(xc,yc); It turns out that it works even if the particle is not added to the ROI Manager but has a record in the results table (you might think the "getResult" would have been an obvious clue for me. Doh!) My current solution doesn't even use the ROI Manager now. (Whew!) In case you are interested, I want to determine the smallest distance between particle perimeters - if two or more are close enough then I want to consider them a single object (like a chain of pearls). I think this macro ... http://www.tidewater.net/~pemaquid/FITVIS_neighbors.ijm ... shows what I mean. Thanks again for all the help! Cheers, Ben Ben Tupper |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by BenTupper
Ben Tupper a écrit :
> On Apr 17, 2009, at 4:08 AM, Senger Fabrice wrote: >>> >> Hi there, >> >> i'm sorry to not have the answer to your problem but I take the >> opportunity to talk about a problem I have. I'm segmenting cells and >> add them to the manager what I would do next is extract each cell to >> generate a new image by using the corresponding ROI. >> Did someone allready tried such an approach ? >> >> Thanks, >> >> Fabrice. >> >> > > > Hi Fabrice, > > If you want to transfer a region from one image to another then you > could do it with something like the following (note that ROI Manager > is not used). > > //START MACRO > > run("Blobs (25K)"); > setAutoThreshold(); > //run("Threshold..."); > setThreshold(125, 255); > run("Convert to Mask"); > maskID = getImageID(); > run("Set Measurements...", "area mean standard modal min centroid > center perimeter bounding fit shape feret's integrated median skewness > kurtosis redirect=None decimal=3"); > run("Analyze Particles...", "size=0-Infinity circularity=0.00-1.00 > show=Nothing display clear record "); > > //first method creates a new image to accept the subimage > ix = 22; > bx = getResult("BX", ix); > by = getResult("BY", ix); > w = getResult("Width", ix); > h = getResult("Height", ix); > > name = "#" + ix+1; > newImage(name, "8-bit White", w,h,1); > img23ID = getImageID(); > > selectImage(maskID); > makeRectangle(bx,by,w,h); > run("Copy"); > selectImage(img23ID); > run("Paste"); > run("Invert"); > > > // the second method exploits the File > New > Internal Clipboard menu > item > ix = 30; > bx = getResult("BX", ix); > by = getResult("BY", ix); > w = getResult("Width", ix); > h = getResult("Height", ix); > > selectImage(maskID); > makeRectangle(bx,by,w,h); > run("Copy"); > name = "#" + ix+1; > run("Internal Clipboard"); > rename(name); > > //END MACRO > > > > Ben Tupper your mail some how gives me a first answer but actually my question is more like : I have a picture with 10 cells in it. I'm able to segment them. Next for each cell I would create a new picture. So I don't want to copy one specific particle, I would like to create one image for each particle. This is part of an analysis scheme where I plan to isolate cells from a picture into a folder and then batch process this folder with another macro. -- Senger Fabrice |
Re: Disjoint ROI selection using macro language
|
|
On Friday 17 April 2009, Senger Fabrice wrote:
> I have a picture with 10 cells in it. > I'm able to segment them. > Next for each cell I would create a new picture. > So I don't want to copy one specific particle, I would like to create > one image for each particle. > This is part of an analysis scheme where I plan to isolate cells from a > picture into a folder and then batch process this folder with another > macro. Have a look at the Catalogue_Particles macro in the Morphology collection. It can create a stack of objects extracted from a binary image (one object per slice). Cheers, G |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by BenTupper
Hi Ben,
just a short remark: If you want to combine particles closer than a given distance, create a mask and dilate it. If you use the Process>Filters>Maximum for dilating, it won't be a square but a circular kernel. For large distances, thresholding the distance map of the inverted mask will be faster than 'max'. Set the threshold range to 0...radius. Michael ________________________________________________________________ On 17 Apr 2009, at 15:02, Ben Tupper wrote: > In case you are interested, I want to determine the smallest > distance between particle perimeters - if two or more are close > enough then I want to consider them a single object (like a chain > of pearls). I think this macro ... |
|
|
Great ideas! Thanks!
On Apr 17, 2009, at 9:44 AM, Michael Schmid wrote: > Hi Ben, > > just a short remark: > If you want to combine particles closer than a given distance, > create a mask and dilate it. If you use the Process>Filters>Maximum > for dilating, it won't be a square but a circular kernel. > > For large distances, thresholding the distance map of the inverted > mask will be faster than 'max'. Set the threshold range to 0...radius. > > Michael > ________________________________________________________________ > > On 17 Apr 2009, at 15:02, Ben Tupper wrote: > >> In case you are interested, I want to determine the smallest >> distance between particle perimeters - if two or more are close >> enough then I want to consider them a single object (like a chain >> of pearls). I think this macro ... Ben Tupper |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by Gabriel Landini
Gabriel Landini a écrit :
> On Friday 17 April 2009, Senger Fabrice wrote: > >> I have a picture with 10 cells in it. >> I'm able to segment them. >> Next for each cell I would create a new picture. >> So I don't want to copy one specific particle, I would like to create >> one image for each particle. >> This is part of an analysis scheme where I plan to isolate cells from a >> picture into a folder and then batch process this folder with another >> macro. >> > > Have a look at the Catalogue_Particles macro in the Morphology collection. > It can create a stack of objects extracted from a binary image (one object per > slice). > > Cheers, > G > i will try again. Actually I wanted to adapt this code taken from : // ROI Manager Stack Demo // n = roiManager("count"); for (i=0; i<n; i++) { roiManager("select", i); getStatistics(area, mean); setResult("Area", i, area); setResult("Mean", i, mean); } updateResults(); roiManager("deselect"); and instead of making measurements I would rather go for a duplicate image command I tried different ways but it does not work yet, one of my macros get stuck to the first ROI and another version only displays the last particle... To me the advantage of using the ROI manager is to vizualize the selected particles and to have the opportunity to discard not relevant particles. Fabrice. -- Senger Fabrice |
Re: Disjoint ROI selection using macro language
|
|
On Friday 17 April 2009, Senger Fabrice wrote:
> I tried this but did not succeeded yet as i get some error messages, but > i will try again. > Actually I wanted to adapt this code taken from : What error do you get? G |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by Fabrice Senger
Hi Fabrice,
what about something like this: myID = getImageID(); n = roiManager("count"); for (i=0; i<n; i++) { selectImage(myID); roiManager("select", i); run("Duplicate...", "title=particle_"+i); } Michael ________________________________________________________________ On 17 Apr 2009, at 17:06, Senger Fabrice wrote: > Gabriel Landini a écrit : >> On Friday 17 April 2009, Senger Fabrice wrote: >> >>> I have a picture with 10 cells in it. >>> I'm able to segment them. >>> Next for each cell I would create a new picture. >>> So I don't want to copy one specific particle, I would like to >>> create >>> one image for each particle. >>> This is part of an analysis scheme where I plan to isolate cells >>> from a >>> picture into a folder and then batch process this folder with >>> another >>> macro. >>> >> >> Have a look at the Catalogue_Particles macro in the Morphology >> collection. >> It can create a stack of objects extracted from a binary image >> (one object per slice). >> >> Cheers, >> G >> > I tried this but did not succeeded yet as i get some error > messages, but i will try again. > Actually I wanted to adapt this code taken from : > > // ROI Manager Stack Demo > // > > n = roiManager("count"); > for (i=0; i<n; i++) { > roiManager("select", i); > getStatistics(area, mean); > setResult("Area", i, area); > setResult("Mean", i, mean); > } > updateResults(); > roiManager("deselect"); > > > and instead of making measurements I would rather go for a > duplicate image command I tried different ways but it does not work > yet, one of my macros get stuck to the first ROI and another > version only displays the last particle... > To me the advantage of using the ROI manager is to vizualize the > selected particles and to have the opportunity to discard not > relevant particles. > > Fabrice. > > -- > Senger Fabrice |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by Fabrice Senger
I think that I found what error you might have encountered. This would happen
if you were using the built in Particle analyzer instead of the Particels8. I uploaded a new version of the morphology plugins that fixes this. http://www.dentistry.bham.ac.uk/landinig/software/morphology.zip Let me know if that fixes it. Cheers G |
Re: Disjoint ROI selection using macro language
|
|
In reply to this post by Michael Schmid
Michael Schmid a écrit :
> Hi Fabrice, > > what about something like this: > > myID = getImageID(); > n = roiManager("count"); > for (i=0; i<n; i++) { > selectImage(myID); > roiManager("select", i); > run("Duplicate...", "title=particle_"+i); > } > > > Michael > ________________________________________________________________ > > On 17 Apr 2009, at 17:06, Senger Fabrice wrote: > >> Gabriel Landini a écrit : >>> On Friday 17 April 2009, Senger Fabrice wrote: >>> >>>> I have a picture with 10 cells in it. >>>> I'm able to segment them. >>>> Next for each cell I would create a new picture. >>>> So I don't want to copy one specific particle, I would like to create >>>> one image for each particle. >>>> This is part of an analysis scheme where I plan to isolate cells >>>> from a >>>> picture into a folder and then batch process this folder with another >>>> macro. >>>> >>> >>> Have a look at the Catalogue_Particles macro in the Morphology >>> collection. >>> It can create a stack of objects extracted from a binary image (one >>> object per slice). >>> >>> Cheers, >>> G >>> >> I tried this but did not succeeded yet as i get some error messages, >> but i will try again. >> Actually I wanted to adapt this code taken from : >> >> // ROI Manager Stack Demo >> // >> >> n = roiManager("count"); >> for (i=0; i<n; i++) { >> roiManager("select", i); >> getStatistics(area, mean); >> setResult("Area", i, area); >> setResult("Mean", i, mean); >> } >> updateResults(); >> roiManager("deselect"); >> >> >> and instead of making measurements I would rather go for a duplicate >> image command I tried different ways but it does not work yet, one of >> my macros get stuck to the first ROI and another version only >> displays the last particle... >> To me the advantage of using the ROI manager is to vizualize the >> selected particles and to have the opportunity to discard not >> relevant particles. >> >> Fabrice. >> >> -- >> Senger Fabrice it works just great, my mistake was in taht simple line run("Duplicate...", "title=particle_"+i); where I tried title=particle_i anyway, thank you very much, Fabrice. I will try aswell Gabriel's suggestion... -- Senger Fabrice |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |