Flow tracking
Locked
 
|
This post was updated on .
  Dear imageJ users, I have recently known about ImageJ and was trying to use it for our medical engineering problem. As you can see in the figures the fluid flows through the serpentine shape channel, see the advance from one figure to the other on the upper serpentine. We need to capture flow motion through the channel from video sequence, by automatically tracking position or speed. Unfortunately I have not found any plugging for this purpose, I have tried several tracking pluggings but they mosly work with particles and do not detect fluid flow. I will strongly appreciate any tip that may help us. Thanks in advance. Iñaki. |
|
|
isadaba wrote:
> http://www.nabble.com/file/p18052351/2665.bmp > > http://www.nabble.com/file/p18052351/2734.bmp > > Dear imageJ users, > > I have recently known about ImageJ and was trying to use it for our medical > engineering problem. As you can see in the figures the fluid flows through > the serpentine shape channel, see the advance from one figure to the other > on the upper serpentine. We need to capture flow motion through the channel > from video sequence, by automatically tracking position or speed. You might be able to track the advancing front by taking the difference between successive frames. There are plugins that can do this. Then you may be able to use tracking plugins to follow the front. --aryeh -- Aryeh Weiss School of Engineering Bar Ilan University Ramat Gan 52900 Israel Ph: 972-3-5317638 FAX: 972-3-7384050 |
|
|
In reply to this post by isadaba
I would suggest taking advantage of the Red color.
Cover the part of the apparatus that you don't want to track - and take your color photos. Alternatively (if its too late to do this), crop your images such that the only red is from the fluid. separate the image into its RGB components and analyze the red channel. It may be necessary to do some thresholding to get the image exclusively your fluid. If the diameter of the tube is equal in all parts of the image - then you don't really need to track it. Just threshold the image, and sum all the pixels. The distance along the tube and the total number of Red pixels should be related by a constant. isadaba wrote: > http://www.nabble.com/file/p18052351/2665.bmp > > http://www.nabble.com/file/p18052351/2734.bmp > > Dear imageJ users, > > I have recently known about ImageJ and was trying to use it for our medical > engineering problem. As you can see in the figures the fluid flows through > the serpentine shape channel, see the advance from one figure to the other > on the upper serpentine. We need to capture flow motion through the channel > from video sequence, by automatically tracking position or speed. > > Unfortunately I have not found any plugging for this purpose, I have tried > several tracking pluggings but they mosly work with particles and do not > detect fluid flow. > > I will strongly appreciate any tip that may help us. > > Thanks in advance. > Iñaki. -- John K. Alexander, Ph.D. Post-Doctoral Fellow William Green Laboratory University of Chicago Dept. Neurobiology, Pharmacology, and Physiology 947 East 58th Street Abott Hall 402 Chicago, IL 60637 (off) 773-702-9386 (fax) 773-702-3774 [hidden email] |
|
|
In reply to this post by Aryeh Weiss
Many thanks for your advice.
Could you please tell the specific plugin for substracting? It sounds quite interesting. Iñaki
|
|
|
In reply to this post by isadaba
Dear Iñaki,
for analysis of the speed you can use a so called time-space plot (also named kymograph). More detailed information about such an analysis and plugins/makros to do it can be found under: http://www.embl.de/eamnet/html/kymograph.html Cheers Arne ____________________________________________________ Dr. Arne Seitz, Scientific Officer Advanced Light Microscopy Facility EMBL Heidelberg +49 6221 387 8467 -----Ursprüngliche Nachricht----- Von: ImageJ Interest Group [mailto:[hidden email]] Im Auftrag von isadaba Gesendet: Sonntag, 22. Juni 2008 19:32 An: [hidden email] Betreff: Flow tracking http://www.nabble.com/file/p18052351/2665.bmp http://www.nabble.com/file/p18052351/2734.bmp Dear imageJ users, I have recently known about ImageJ and was trying to use it for our medical engineering problem. As you can see in the figures the fluid flows through the serpentine shape channel, see the advance from one figure to the other on the upper serpentine. We need to capture flow motion through the channel from video sequence, by automatically tracking position or speed. Unfortunately I have not found any plugging for this purpose, I have tried several tracking pluggings but they mosly work with particles and do not detect fluid flow. I will strongly appreciate any tip that may help us. Thanks in advance. Iñaki. -- View this message in context: http://www.nabble.com/Flow-tracking-tp18052351p18052351.html Sent from the ImageJ mailing list archive at Nabble.com. |
Playback of large movies with high frame rates
|
|
In reply to this post by isadaba
Dear ImageJ listserver,
I have a small problem I was hoping the community might be able to advise me on. I have several long time-lapse live-cell image sequences captured using TIRF microscopy. 1000 images were captured with a 200 ms exposure time and a 500 ms delay in between exposures. A typical image sequence file size is about 500 MB. It is necessary to capture many images and with this rate of repetition since the the cells I am imaging express a fluorescent membrane protein that exhibits very interesting dynamics (both fast and slow motions) and over a long time period. I do not have the option of capturing fewer images over the same time interval since doing so will cause the software to miss recording very transient and fast vesicle fusion events with the membrane. We use ImageJ for analysis and we run the program on a powerful Linux computer with 4GB of RAM. To see the protein dynamics, it is necessary to play back the image sequence with a 300 ms frame rate on the computer screen in ImageJ, which is possible with the latest version of ImageJ (max frame rate 1000 fps or frame rate allowed by memory and processor speed). If played slower than this, then it's difficult for the human eye to discern a fusion event and notice any protein motion. My problem is that I would like to make these movies portable for presentations (in Powerpoint or Quicktime for example) without too much compression that masks the features I'm trying to show and without slowing them down to a lower frame rate. I usually save image sequences as .avi files, but I've noticed that if I do this when the display frame rate in ImageJ is greater than 100 ms, the avi is saved with a frame rate of 100 ms. Is this a bug or is it simply a limitation of the .avi format, or is it a limitation of the player I use to playback the .avi file (Quicktime)? I can find no mention of a maximum frame rate in the .avi writer plugin on the ImageJ website. Even if I can save files properly with these high frame rates, I likely will not be able to embed them in Powerpoint and play them at that speed since the demands on my laptop for this file size and display rate are too great. Does anyone know of an alternative way for displaying large movies with high frame rates? Thank you for your time. John Oreopoulos, BSc, PhD Candidate University of Toronto Institute For Biomaterials and Biomedical Engineering Centre For Studies in Molecular Imaging Tel: W:416-946-5022 |
Re: Playback of large movies with high frame rates
|
|
My mistake, I should have written "fps" for the playback frame rate
instead of "ms". I got my units mixed up. The sentences should read: To see the protein dynamics, it is necessary to play back the image sequence with a 300 fps frame rate on the computer screen in ImageJ, which is possible with the latest version of ImageJ (max frame rate 1000 fps or frame rate allowed by memory and processor speed). I usually save image sequences as .avi files, but I've noticed that if I do this when the display frame rate in ImageJ is greater than 100 fps, the avi is saved with a frame rate of 100 fps. John On 23-Jun-08, at 2:27 PM, John Oreopoulos wrote: > Dear ImageJ listserver, > > I have a small problem I was hoping the community might be able to > advise me on. I have several long time-lapse live-cell image > sequences captured using TIRF microscopy. 1000 images were captured > with a 200 ms exposure time and a 500 ms delay in between > exposures. A typical image sequence file size is about 500 MB. It > is necessary to capture many images and with this rate of > repetition since the the cells I am imaging express a fluorescent > membrane protein that exhibits very interesting dynamics (both fast > and slow motions) and over a long time period. I do not have the > option of capturing fewer images over the same time interval since > doing so will cause the software to miss recording very transient > and fast vesicle fusion events with the membrane. We use ImageJ for > analysis and we run the program on a powerful Linux computer with > 4GB of RAM. To see the protein dynamics, it is necessary to play > back the image sequence with a 300 ms frame rate on the computer > screen in ImageJ, which is possible with the latest version of > ImageJ (max frame rate 1000 fps or frame rate allowed by memory and > processor speed). If played slower than this, then it's difficult > for the human eye to discern a fusion event and notice any protein > motion. > My problem is that I would like to make these movies portable for > presentations (in Powerpoint or Quicktime for example) without too > much compression that masks the features I'm trying to show and > without slowing them down to a lower frame rate. > I usually save image sequences as .avi files, but I've noticed that > if I do this when the display frame rate in ImageJ is greater than > 100 ms, the avi is saved with a frame rate of 100 ms. Is this a bug > or is it simply a limitation of the .avi format, or is it a > limitation of the player I use to playback the .avi file > (Quicktime)? I can find no mention of a maximum frame rate in > the .avi writer plugin on the ImageJ website. > Even if I can save files properly with these high frame rates, I > likely will not be able to embed them in Powerpoint and play them > at that speed since the demands on my laptop for this file size and > display rate are too great. Does anyone know of an alternative way > for displaying large movies with high frame rates? > > Thank you for your time. > > > John Oreopoulos, BSc, > PhD Candidate > University of Toronto > Institute For Biomaterials and Biomedical Engineering > Centre For Studies in Molecular Imaging > > Tel: W:416-946-5022 |
|
|
In reply to this post by isadaba
You can start with the "stacks-T functions" plugins which are part of the WCIF
plugins collection (or is is now called MBF?). I think that deltaF-up and deltaF-down may do it. If you need to do it yourself, you can look in the example macros on the ij website. For example, StackProfilePlot will show you how to loop through your stack. The you can take slices n and n-1 (or n+1), subtract them, and put them in a new stack. Look at the "SubtractMeasuredBackground" macro on the in http://rsb.info.nih.gov/ij/macros/SubractMeasuredBackground.txt It is close to what you want -- just change the constant background to be the previous (or next) slice. --aryeh isadaba wrote: > Many thanks for your advice. > Could you please tell the specific plugin for substracting? > > It sounds quite interesting. > Iñaki > > > > > Aryeh Weiss wrote: >> isadaba wrote: >>> http://www.nabble.com/file/p18052351/2665.bmp >>> >>> http://www.nabble.com/file/p18052351/2734.bmp >>> >>> Dear imageJ users, >>> >>> I have recently known about ImageJ and was trying to use it for our >>> medical >>> engineering problem. As you can see in the figures the fluid flows >>> through >>> the serpentine shape channel, see the advance from one figure to the >>> other >>> on the upper serpentine. We need to capture flow motion through the >>> channel >>> from video sequence, by automatically tracking position or speed. >> You might be able to track the advancing front by taking the difference >> between >> successive frames. There are plugins that can do this. Then you may be >> able to >> use tracking plugins to follow the front. >> >> --aryeh >> -- >> Aryeh Weiss >> School of Engineering >> Bar Ilan University >> Ramat Gan 52900 Israel >> >> Ph: 972-3-5317638 >> FAX: 972-3-7384050 >> >> > |
Re: Playback of large movies with high frame rates
|
|
In reply to this post by John Oreopoulos
Frame rate is stored in the AVI RIFF main header in the field
dwMicroSecPerFrame, so I doubt its a limitation of the file format. Redrawing 300 images per second can be a pretty significant load on the system, though, not to mention beyond the refresh rate of most monitors by about 5x. If the motion you've captured is as slow as you describe, perhaps you could save something like every 5th or 10th frame of your original movie and output an AVI at either 60 or 30 fps? If you've been viewing the videos in ImageJ at "300 fps," you've probably been doing essentially that since your monitor most likely works in the 60 Hz range (not to mention the limitations of the human eye). Charlie Shaw Graduate Student Macromolecular Science & Engineering Program University of Michigan, Ann Arbor On Jun 23, 2008, at 2:57 PM, John Oreopoulos wrote: > My mistake, I should have written "fps" for the playback frame rate > instead of "ms". I got my units mixed up. The sentences should read: > > To see the protein dynamics, it is necessary to play back the image > sequence with a 300 fps frame rate on the computer screen in ImageJ, > which is possible with the latest version of ImageJ (max frame rate > 1000 fps or frame rate allowed by memory and processor speed). > > I usually save image sequences as .avi files, but I've noticed that if > I do this when the display frame rate in ImageJ is greater than 100 > fps, the avi is saved with a frame rate of 100 fps. > > John > > On 23-Jun-08, at 2:27 PM, John Oreopoulos wrote: > >> Dear ImageJ listserver, >> >> I have a small problem I was hoping the community might be able to >> advise me on. I have several long time-lapse live-cell image >> sequences captured using TIRF microscopy. 1000 images were captured >> with a 200 ms exposure time and a 500 ms delay in between exposures. >> A typical image sequence file size is about 500 MB. It is necessary >> to capture many images and with this rate of repetition since the the >> cells I am imaging express a fluorescent membrane protein that >> exhibits very interesting dynamics (both fast and slow motions) and >> over a long time period. I do not have the option of capturing fewer >> images over the same time interval since doing so will cause the >> software to miss recording very transient and fast vesicle fusion >> events with the membrane. We use ImageJ for analysis and we run the >> program on a powerful Linux computer with 4GB of RAM. To see the >> protein dynamics, it is necessary to play back the image sequence >> with a 300 ms frame rate on the computer screen in ImageJ, which is >> possible with the latest version of ImageJ (max frame rate 1000 fps >> or frame rate allowed by memory and processor speed). If played >> slower than this, then it's difficult for the human eye to discern a >> fusion event and notice any protein motion. >> My problem is that I would like to make these movies portable for >> presentations (in Powerpoint or Quicktime for example) without too >> much compression that masks the features I'm trying to show and >> without slowing them down to a lower frame rate. >> I usually save image sequences as .avi files, but I've noticed that >> if I do this when the display frame rate in ImageJ is greater than >> 100 ms, the avi is saved with a frame rate of 100 ms. Is this a bug >> or is it simply a limitation of the .avi format, or is it a >> limitation of the player I use to playback the .avi file (Quicktime)? >> I can find no mention of a maximum frame rate in the .avi writer >> plugin on the ImageJ website. >> Even if I can save files properly with these high frame rates, I >> likely will not be able to embed them in Powerpoint and play them at >> that speed since the demands on my laptop for this file size and >> display rate are too great. Does anyone know of an alternative way >> for displaying large movies with high frame rates? >> >> Thank you for your time. >> >> >> John Oreopoulos, BSc, >> PhD Candidate >> University of Toronto >> Institute For Biomaterials and Biomedical Engineering >> Centre For Studies in Molecular Imaging >> >> Tel: W:416-946-5022 > |
Re: Playback of large movies with high frame rates
|
|
Hi John,
Since ImageJ version 1.40f, Image>Stackes>Animation Options has a limit of 1000 frames per second. Nevertheless, the current version of ImageJ limits the frame rate of an AVI file to a maximum of 60 frames per second. I guess that this is to avoid problems with AVI players that may have problems with higher frame rates (they are written for cinema movies, not for scientific data...). Also, in most cases, very high frame rates in an AVI file are not useful since hardware (monitors, graphics cards) cannot display hundreds of frames per second. Some players (e.g. VLC) have an option to change the playback rate (time-lapse), so you could use this for quickly browsing through the movie. If you want to try creating an avi with higher frame rate: Download the current version of the AVI_Writer sources from http://rsb.info.nih.gov/ij/source/ij/plugin/filter/AVI_Writer.java replace the first line package ij.plugin.filter; by import ij.plugin.filter.*; and modify the method private double getFrameRate(ImagePlus imp) by replacing "60.0" with whatever maximum frame rate you like. Then you can use it as an external plugin with "Compile and run" Hope this helps, Michael ________________________________________________________________ On 24 Jun 2008, at 14:00, Charles Shaw wrote: > Frame rate is stored in the AVI RIFF main header in the field > dwMicroSecPerFrame, so I doubt its a limitation of the file > format. Redrawing 300 images per second can be a pretty > significant load on the system, though, not to mention beyond the > refresh rate of most monitors by about 5x. > > If the motion you've captured is as slow as you describe, perhaps > you could save something like every 5th or 10th frame of your > original movie and output an AVI at either 60 or 30 fps? If you've > been viewing the videos in ImageJ at "300 fps," you've probably > been doing essentially that since your monitor most likely works in > the 60 Hz range (not to mention the limitations of the human eye). > > Charlie Shaw > Graduate Student > Macromolecular Science & Engineering Program > University of Michigan, Ann Arbor > > On Jun 23, 2008, at 2:57 PM, John Oreopoulos wrote: > >> My mistake, I should have written "fps" for the playback frame >> rate instead of "ms". I got my units mixed up. The sentences >> should read: >> >> To see the protein dynamics, it is necessary to play back the >> image sequence with a 300 fps frame rate on the computer screen in >> ImageJ, which is possible with the latest version of ImageJ (max >> frame rate 1000 fps or frame rate allowed by memory and processor >> speed). >> >> I usually save image sequences as .avi files, but I've noticed >> that if I do this when the display frame rate in ImageJ is greater >> than 100 fps, the avi is saved with a frame rate of 100 fps. >> >> John >> >> On 23-Jun-08, at 2:27 PM, John Oreopoulos wrote: >> >>> Dear ImageJ listserver, >>> >>> I have a small problem I was hoping the community might be able >>> to advise me on. I have several long time-lapse live-cell image >>> sequences captured using TIRF microscopy. 1000 images were >>> captured with a 200 ms exposure time and a 500 ms delay in >>> between exposures. A typical image sequence file size is about >>> 500 MB. It is necessary to capture many images and with this rate >>> of repetition since the the cells I am imaging express a >>> fluorescent membrane protein that exhibits very interesting >>> dynamics (both fast and slow motions) and over a long time >>> period. I do not have the option of capturing fewer images over >>> the same time interval since doing so will cause the software to >>> miss recording very transient and fast vesicle fusion events with >>> the membrane. We use ImageJ for analysis and we run the program >>> on a powerful Linux computer with 4GB of RAM. To see the protein >>> dynamics, it is necessary to play back the image sequence with a >>> 300 ms frame rate on the computer screen in ImageJ, which is >>> possible with the latest version of ImageJ (max frame rate 1000 >>> fps or frame rate allowed by memory and processor speed). If >>> played slower than this, then it's difficult for the human eye to >>> discern a fusion event and notice any protein motion. >>> My problem is that I would like to make these movies portable for >>> presentations (in Powerpoint or Quicktime for example) without >>> too much compression that masks the features I'm trying to show >>> and without slowing them down to a lower frame rate. >>> I usually save image sequences as .avi files, but I've noticed >>> that if I do this when the display frame rate in ImageJ is >>> greater than 100 ms, the avi is saved with a frame rate of 100 >>> ms. Is this a bug or is it simply a limitation of the .avi >>> format, or is it a limitation of the player I use to playback >>> the .avi file (Quicktime)? I can find no mention of a maximum >>> frame rate in the .avi writer plugin on the ImageJ website. >>> Even if I can save files properly with these high frame rates, I >>> likely will not be able to embed them in Powerpoint and play them >>> at that speed since the demands on my laptop for this file size >>> and display rate are too great. Does anyone know of an >>> alternative way for displaying large movies with high frame rates? >>> >>> Thank you for your time. >>> >>> >>> John Oreopoulos, BSc, >>> PhD Candidate >>> University of Toronto >>> Institute For Biomaterials and Biomedical Engineering >>> Centre For Studies in Molecular Imaging >>> >>> Tel: W:416-946-5022 >> |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |