Hough transform question
Locked
 
|
Hi,
We have a series of electron microscope images of spherical viruses that have been treated in different ways. The images of the untreated virus show (potentially overlapping) circular objects (dark against a light background). We suspect that various treatments are causing a structural deformation and an initial question that we would like answered is whether the treatments are affecting how "circular" the virus particles are. I know that Hough transforms can be used to identify circular objects and I suspect that approach could be used to identify the individual virus particles in these images. What I'd like to know is whether there is then some way to quantify the "circularity" of them. The visible deviations from circularity can be as pronounced as irregular shapes or as subtle as a transformation into slightly oval shapes. The treatments can also cause an increase in "broken" virus particles (incomplete circles, for example). We would like to be able to quantify these effects. Any advice would be welcome. -- David Gene Morgan Electron Microscopy Center 047D Simon Hall IU Bloomington 812 856 1457 (office) 812 856 3221 (3200) http://bio.indiana.edu/~cryo -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Hi David,
Have you thought of using image moments for the circularity? If they are distorted, the ratio of the principal axes might be a good measure. If your background is pretty uniform, this might work. Otherwise, the background might influence things too much. Hope this helps... Michael On Mon, Aug 3, 2015 at 9:08 AM David Gene Morgan <[hidden email]> wrote: > Hi, > > We have a series of electron microscope images of spherical viruses > that have been treated in different ways. The images of the untreated > virus show (potentially overlapping) circular objects (dark against a > light background). We suspect that various treatments are causing a > structural deformation and an initial question that we would like > answered is whether the treatments are affecting how "circular" the > virus particles are. > > I know that Hough transforms can be used to identify circular > objects > and I suspect that approach could be used to identify the individual > virus particles in these images. What I'd like to know is whether there > is then some way to quantify the "circularity" of them. > > The visible deviations from circularity can be as pronounced as > irregular shapes or as subtle as a transformation into slightly oval > shapes. The treatments can also cause an increase in "broken" virus > particles (incomplete circles, for example). We would like to be able > to quantify these effects. > > Any advice would be welcome. > > -- > David Gene Morgan > Electron Microscopy Center > 047D Simon Hall > IU Bloomington > 812 856 1457 (office) > 812 856 3221 (3200) > http://bio.indiana.edu/~cryo > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Re: Hough transform question
|
|
In reply to this post by David Gene Morgan
David,
Could you please provide a link to sample images? Thx e.- -----Original Message----- From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Michael Sarahan Sent: Monday, August 03, 2015 1:06 PM To: [hidden email] Subject: Re: Hough transform question Hi David, Have you thought of using image moments for the circularity? If they are distorted, the ratio of the principal axes might be a good measure. If your background is pretty uniform, this might work. Otherwise, the background might influence things too much. Hope this helps... Michael On Mon, Aug 3, 2015 at 9:08 AM David Gene Morgan <[hidden email]> wrote: > Hi, > > We have a series of electron microscope images of spherical > viruses that have been treated in different ways. The images of the > untreated virus show (potentially overlapping) circular objects (dark > against a light background). We suspect that various treatments are > causing a structural deformation and an initial question that we would > like answered is whether the treatments are affecting how "circular" > the virus particles are. > > I know that Hough transforms can be used to identify circular > objects and I suspect that approach could be used to identify the > individual virus particles in these images. What I'd like to know is > whether there is then some way to quantify the "circularity" of them. > > The visible deviations from circularity can be as pronounced > as irregular shapes or as subtle as a transformation into slightly > oval shapes. The treatments can also cause an increase in "broken" > virus particles (incomplete circles, for example). We would like to > be able to quantify these effects. > > Any advice would be welcome. > > -- > David Gene Morgan > Electron Microscopy Center > 047D Simon Hall > IU Bloomington > 812 856 1457 (office) > 812 856 3221 (3200) > http://bio.indiana.edu/~cryo > > -- > 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 |
Re: Hough transform question
|
|
Hi,
Thanks. I had to check to make sure that I could post any of the images. Here are links to a pair of them: https://iu.box.com/s/l1x1d3ckcya8kmmi3hbtwyqldry5m0nm https://iu.box.com/s/g85sj3orr5i4z1d62brn9mxrp4pdv60t On 08/07/2015 09:36 AM, Kischell, Eric R., M.S. wrote: > David, > Could you please provide a link to sample images? > Thx e.- > > > -----Original Message----- > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Michael Sarahan > Sent: Monday, August 03, 2015 1:06 PM > To: [hidden email] > Subject: Re: Hough transform question > > Hi David, > > Have you thought of using image moments for the circularity? If they are distorted, the ratio of the principal axes might be a good measure. If your background is pretty uniform, this might work. Otherwise, the background might influence things too much. > > Hope this helps... > Michael > > On Mon, Aug 3, 2015 at 9:08 AM David Gene Morgan <[hidden email]> > wrote: > >> Hi, >> >> We have a series of electron microscope images of spherical >> viruses that have been treated in different ways. The images of the >> untreated virus show (potentially overlapping) circular objects (dark >> against a light background). We suspect that various treatments are >> causing a structural deformation and an initial question that we would >> like answered is whether the treatments are affecting how "circular" >> the virus particles are. >> >> I know that Hough transforms can be used to identify circular >> objects and I suspect that approach could be used to identify the >> individual virus particles in these images. What I'd like to know is >> whether there is then some way to quantify the "circularity" of them. >> >> The visible deviations from circularity can be as pronounced >> as irregular shapes or as subtle as a transformation into slightly >> oval shapes. The treatments can also cause an increase in "broken" >> virus particles (incomplete circles, for example). We would like to >> be able to quantify these effects. >> >> Any advice would be welcome. >> >> -- >> David Gene Morgan >> Electron Microscopy Center >> 047D Simon Hall >> IU Bloomington >> 812 856 1457 (office) >> 812 856 3221 (3200) >> http://bio.indiana.edu/~cryo >> >> -- >> 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 > -- David Gene Morgan Electron Microscopy Center 047D Simon Hall IU Bloomington 812 856 1457 (office) 812 856 3221 (3200) http://bio.indiana.edu/~cryo -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
Batch mode and displaying images
|
|
Hi all,
I like to use batch mode where i can in macros to speed things up. Sometimes though i want to interact with image windows at a certain point in the macro and i have been using setBatchMode("Exit and display") and then i set batch mode true once the image interaction is complete. However this seems to run very slowly. I know you're not really supposed to use multiple instances of setBatchMode(true). I thought that setBatchMode("show") might be the solution. I run a loop to show 3 images and the active image is the last one in the loop, i can add a selection to this image but not to the others that were shown earlier within the loop. This suggests i have to use setBatchMode("show") and do the interaction step on each image one by one sequentially instead of showing them all first and then adding selections. Whats the best way to use batch mode when you want to interact with images as well? Thanks for the advice, Matt -- Matt Pearson Microscopy Facility MRC Human Genetics Unit IGMM University of Edinburgh Crewe Road EH4 2XU -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |