Turbulent Jet Morphology?

[Frank Shaffer]

Attached is an image from a high-speed video showing turbulence eddies of a submerged jet (UC Berkeley Fluid Dynamics Lab).  We would like to characterize the morphology (sizes, length scales, etc.).  

Any suggestions for tools to use in ImageJ?

Thank you,
Frank Shaffer


--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Herbie]

Good day Frank,

This is a really impressive image...

It looks like a case for regional fractal analyses, though I'm not an expert
in this field.

<http://imagej.1557.x6.nabble.com/file/t380516/Test_005_Fluorescent_Dye_enhanced_UCB-FML-Y3.png>
Attached please find a binary version of the central 512x512 image portion.
I've first applied an ImageJ-bandpass filter with parameters 40, 10 and then
used the automatic threshold scheme "MaxEntropy".
(The binarized result strongly depends on the filter parameters and the
threshold scheme. It is up to you to determine the dependency of the fractal
analyses on theses parameters. If there isn't a great dependency, the
approach may be useful.)

The binary image may serve for fractal analyses such as "Box Counting" etc.
<https://en.wikipedia.org/wiki/Box_counting>

HTH

Herbie

:::::::::::::::::::::::::::::::::::::::::::

Frank Shaffer wrote

> Attached is an image from a high-speed video showing turbulence eddies of
> a submerged jet (UC Berkeley Fluid Dynamics Lab).  We would like to
> characterize the morphology (sizes, length scales, etc.).  
>
> Any suggestions for tools to use in ImageJ?
>
> Thank you,
> Frank Shaffer
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
>
> Test_005_Fluorescent_Dye_enhanced_UCB-FML-Y3.2-S1 Camera.jpg (270K)
> &lt;http://imagej.1557.x6.nabble.com/attachment/5022840/0/Test_005_Fluorescent_Dye_enhanced_UCB-FML-Y3.2-S1%20Camera.jpg&gt;

--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Gary Settles]

Hello Frank,

Herbie’s suggestion is a good one. I’ll make a few more, based on my
experience using ImageJ to analyze turbulent jets (line-of-sight-integrated
optical images, though, not fluorescence).
1) Load your video as an image stack into ImageJ. Draw a line on the image
stack indicating the location of the line of pixels to be resliced (e.g. jet
centerline). Then use Image Stacks Reslice followed by keyboard button “/”.
What you get is called, in ImageJ, a “pseudo-linescan” image, but note that
the x,y,z stack in ImageJ is actually an x,y,t stack, where t is the
timeline of your video. With proper calibration this streak image will give
you eddy velocity data.
2) You can see the FFT of an entire image using Process FFT, but a better
way (for me) is to use the plugin “nr realft”, which needs a 32-bit
grayscale image with a selected line segment of length = a power of 2. This
yields a power spectrum plot along the selected line. With some experience
you’ll see that there is a lot more that ImageJ can do with turbulent flows.

Gary Settles
Dist. Prof. Emeritus of M. E., Penn State




--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Frank Shaffer]

Hello Professor Settles,
Thank you for the suggestions. I certainly will try them.
I don't know if you remember, but I worked at a DOE lab in Pittsburgh.  If
I recall correctly, you also were using a copper-vapor laser for flow viz.
Boy were they a headache to use! Now it can all be done with LED's.
I'm now working with Professor Savas at UC Berkeley.
Best regards,
Frank

On Sun, Jan 19, 2020 at 12:40 PM Gary Settles <[hidden email]> wrote:

> Hello Frank,
>
> Herbie’s suggestion is a good one. I’ll make a few more, based on my
> experience using ImageJ to analyze turbulent jets (line-of-sight-integrated
> optical images, though, not fluorescence).
> 1) Load your video as an image stack into ImageJ. Draw a line on the image
> stack indicating the location of the line of pixels to be resliced (e.g.
> jet
> centerline). Then use Image Stacks Reslice followed by keyboard button “/”.
> What you get is called, in ImageJ, a “pseudo-linescan” image, but note that
> the x,y,z stack in ImageJ is actually an x,y,t stack, where t is the
> timeline of your video. With proper calibration this streak image will give
> you eddy velocity data.
> 2) You can see the FFT of an entire image using Process FFT, but a better
> way (for me) is to use the plugin “nr realft”, which needs a 32-bit
> grayscale image with a selected line segment of length = a power of 2. This
> yields a power spectrum plot along the selected line. With some experience
> you’ll see that there is a lot more that ImageJ can do with turbulent
> flows.
>
> Gary Settles
> Dist. Prof. Emeritus of M. E., Penn State
>
>
>
>
> --
> Sent from: http://imagej.1557.x6.nabble.com/
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Frank Shaffer]

Professor Settles,
Thanks for the suggestions! I will try them.
I used to work at a DOE lab in Pittsburgh. I was using a copper-vapor laser for flow viz, and I recall that you were also.  CuV lasers were a headache. Now we can use LED's.

Apparently we can't attach files to replies on ImageJ listserv.  So here are links to some of my videos of our simultaneous schlieren, fluorescent, and PTV imaging of turbulent jets.
www.fdshaffer.net/Playground_Examples.html

Also, a very large submerged jet of jet fuel at
www.FDShaffer.net

I always use ImageJ for all of my flow viz work!  I used Erik Meijering's MTrackJ for the particle trajectories shown in my videos.
Best regards,
Frank Shaffer

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Dimiter Prodanov-3]

Dear Gary,

It looks like a problem for fractal analysis.
You could slice the image between different successive thresholds and then
run the Fractal dimension plugin.
I am not sure that 1 threshold will reveal the morphology because you have
an interplay between illumination effects and perspective.

best regards,

Dimiter

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Jeremy Adler]

The object is three dimensional but the image is two dimensional, effectively a maximum projection.
Fractal analysis really requires a 3D dataset and a3D measurement.

Jeremy Adler
Uppsala U

-----Original Message-----
From: ImageJ Interest Group <[hidden email]> On Behalf Of Dimiter Prodanov
Sent: Monday, January 20, 2020 11:12 AM
To: [hidden email]
Subject: Re: Turbulent Jet Morphology?

Dear Gary,

It looks like a problem for fractal analysis.
You could slice the image between different successive thresholds and then run the Fractal dimension plugin.
I am not sure that 1 threshold will reveal the morphology because you have an interplay between illumination effects and perspective.

best regards,

Dimiter

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html








När du har kontakt med oss på Uppsala universitet med e-post så innebär det att vi behandlar dina personuppgifter. För att läsa mer om hur vi gör det kan du läsa här: http://www.uu.se/om-uu/dataskydd-personuppgifter/

E-mailing Uppsala University means that we will process your personal data. For more information on how this is performed, please read here: http://www.uu.se/en/about-uu/data-protection-policy

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Gabriel Landini]

On Monday, 20 January 2020 10:26:56 GMT [hidden email] wrote:
> The object is three dimensional but the image is two dimensional,
> effectively a maximum projection. Fractal analysis really requires a 3D
> dataset and a3D measurement.

Hi Jeremy,
Not necessarily in every case. You can "section" the fractal, i..e. intersect
the 3D set with a 2D plane and analyse the intersection instead (called the
zerset, then add 1 to the value of the FD result). There is a theorem for
that, but I do not have the reference at hand right now.
I agree that one cannot do that directly from the photo, but maybe they could
illuminate the 3D object with a "laser sheet" to obtain the zeroset that way.

Cheers

Gabriel

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Gabriel Landini]

On Monday, 20 January 2020 10:50:51 GMT I wrote:
> (called the zerset, then add 1 to the value of the FD result). There is a

Sorry I meant "zeroset".

G.

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Herbie]

Good day all,

although the discussion is interesting, especially for me as a "jet and
fractal"-novice, the OP originally asked to specify the provide image, not
3D-sections and the like!

Actually, the method suggested earlier can lead to fractal measures that
change in a meaningful way when applied to regional analyses of the provided
image. As far as I can judge, simple "Box Counting" shows that the binary
regional images are fractal and that the D-measure increases from left to
right in a non-linear fashion.

Regards

Herbie

::::::::::::::::::::::::::::::::::::::::::::::::::::::

Gabriel Landini wrote
> On Monday, 20 January 2020 10:26:56 GMT

> jeremy.adler@.UU

>  wrote:
>> The object is three dimensional but the image is two dimensional,
>> effectively a maximum projection. Fractal analysis really requires a 3D
>> dataset and a3D measurement.
>
> Hi Jeremy,
> Not necessarily in every case. You can "section" the fractal, i..e.
> intersect
> the 3D set with a 2D plane and analyse the intersection instead (called
> the
> zerset, then add 1 to the value of the FD result). There is a theorem for
> that, but I do not have the reference at hand right now.
> I agree that one cannot do that directly from the photo, but maybe they
> could
> illuminate the 3D object with a "laser sheet" to obtain the zeroset that
> way.
>
> Cheers
>
> Gabriel
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Frank Shaffer]

I don't have experience with the fractal recognition technique.
If anyone would like to try their suggestions on an example video, you can download it here:

www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi

It's only 31 frames.

If you technique works well, I will be happy to give you credit in our publications.

Thanks,
Frank Shaffer
FDShaffer.Net

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Herbie]

Dear Frank,

initially you provided a single image and now you provide an image series.
I'm totally unsure what you like to see in the end.

With respect to the single image you wrote:
"We would like to characterize the morphology (sizes, length scales, etc.)."

Is this still what you want or what does the image sequence mean?

If you like, I can provide the result of my initial idea which works for
single images.

BTW I've provided a link to the Wiki dealing with "Box Counting"-

Regards

Herbie

:::::::::::::::::::::::::::::::::::::::::::

Frank Shaffer wrote

> I don't have experience with the fractal recognition technique.
> If anyone would like to try their suggestions on an example video, you can
> download it here:
>
> www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
>
> It's only 31 frames.
>
> If you technique works well, I will be happy to give you credit in our
> publications.
>
> Thanks,
> Frank Shaffer
> FDShaffer.Net
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Frank Shaffer]

Herbie,
The feature recognition only needs to work on one image. Then it can be
repeated for the other images.
If you find something that works, perhaps you can record it in a macro and
post it.
Thanks
Frank

On Mon, Jan 20, 2020, 10:02 AM Herbie <[hidden email]> wrote:

> Dear Frank,
>
> initially you provided a single image and now you provide an image series.
> I'm totally unsure what you like to see in the end.
>
> With respect to the single image you wrote:
> "We would like to characterize the morphology (sizes, length scales,
> etc.)."
>
> Is this still what you want or what does the image sequence mean?
>
> If you like, I can provide the result of my initial idea which works for
> single images.
>
> BTW I've provided a link to the Wiki dealing with "Box Counting"-
>
> Regards
>
> Herbie
>
> :::::::::::::::::::::::::::::::::::::::::::
>
> Frank Shaffer wrote
> > I don't have experience with the fractal recognition technique.
> > If anyone would like to try their suggestions on an example video, you
> can
> > download it here:
> >
> > www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
> >
> > It's only 31 frames.
> >
> > If you technique works well, I will be happy to give you credit in our
> > publications.
> >
> > Thanks,
> > Frank Shaffer
> > FDShaffer.Net
> >
> > --
> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
>
>
>
>
> --
> Sent from: http://imagej.1557.x6.nabble.com/
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Robert Dougherty]

Frank,
Thanks for an interesting question.  It has generated a lot of useful suggestions.
I couldn't stop myself from playing around with the image.  Here is another proposal:
run("Find Ridges", "gaussian_radius=2 minimum_thinning=50");
setOption("BlackBackground", false);
run("Make Binary");
run("Local Thickness (complete process)", "threshold=128 inverse");
run("Add...", "value=15");
run("Reciprocal");
run("Gaussian Blur...", "sigma=30");
run("Reciprocal");
run("Max...", "value=128");
run("Invert LUT");
resetMinAndMax();

Notes:1. The macro is supposed to produce a map of the average size of the vortices.
2. You can download Find Ridges and Local Thickness from https://www.optinav.info/imagej.html  

3. It is important to run this macro on individual slices of your movie.  Doing the whole stack at once would confuse Local Thickness.
4a. The map of the borders between the vortices that appears after Find Ridges might be a good starting point for the other types of analysis that have been mentioned, including reslicing the time stack and perhaps doing a Hough transform to quantify the convection speed.
4b. Do these borders have a name in turbulence research?  Child shear layers?4c. Perhaps you could also find these from 3D PIV data? If so, you use the Local Thickness plugin in 3D mode.


Bob Dougherty

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Herbie]

Dear Frank,

attached (see the follwing link) please find an ImageJ-macro that does the
proposed regional "Box Counting"-analyses for all images of a stack.
fractalDim.ijm
<http://imagej.1557.x6.nabble.com/file/t380516/fractalDim.ijm>  
(Perhaps you start with a single image, because the writing to the
results-table takes a while...)

Paste the macro code to an empty macro window (Plugins >> New >> Macro) and
run it with either a single image or a stack open in ImageJ.

Regional analyses are made for image selections that are 128pel wide.
The increment is half the width, i.e. 64pel, which means an overlap.

The filter parameters are not critical.

The automatic "Otsu"-threshold scheme works quite well for the supplied
image stack.

The D-Column of the results table gives the fractal dimension as defined by
"Box Counting"-approach. See "Examples" in:
<https://en.wikipedia.org/wiki/Fractal_dimension>

Please also see the ImageJ user guide sub-subsection "30.14.2 Fractal Box
Count...":
<https://imagej.nih.gov/ij/docs/guide/146-30.html#toc-Subsection-30.14>

HTH

Herbie

:::::::::::::::::::::::::::::::::::::::::::

Frank Shaffer wrote
> Herbie,
> The feature recognition only needs to work on one image. Then it can be
> repeated for the other images.
> If you find something that works, perhaps you can record it in a macro and
> post it.
> Thanks
> Frank
>
> On Mon, Jan 20, 2020, 10:02 AM Herbie &lt;

> l16@

> &gt; wrote:
>
>> Dear Frank,
>>
>> initially you provided a single image and now you provide an image
>> series.
>> I'm totally unsure what you like to see in the end.
>>
>> With respect to the single image you wrote:
>> "We would like to characterize the morphology (sizes, length scales,
>> etc.)."
>>
>> Is this still what you want or what does the image sequence mean?
>>
>> If you like, I can provide the result of my initial idea which works for
>> single images.
>>
>> BTW I've provided a link to the Wiki dealing with "Box Counting"-
>>
>> Regards
>>
>> Herbie
>>
>> :::::::::::::::::::::::::::::::::::::::::::
>>
>> Frank Shaffer wrote
>> > I don't have experience with the fractal recognition technique.
>> > If anyone would like to try their suggestions on an example video, you
>> can
>> > download it here:
>> >
>> > www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
>> >
>> > It's only 31 frames.
>> >
>> > If you technique works well, I will be happy to give you credit in our
>> > publications.
>> >
>> > Thanks,
>> > Frank Shaffer
>> > FDShaffer.Net
>> >
>> > --
>> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>>
>>
>>
>>
>> --
>> Sent from: http://imagej.1557.x6.nabble.com/
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
Sent from: http://imagej.1557.x6.nabble.com/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Ion Andronache]

[KennethUdut]

I'm new to this group but I've been using ImageJ (Fiji) on a recreational
basis for a couple of years as well as learning what I can about
mathematical morphology. Your animation is a great one and I couldn't help
but play around with it.

I happened to have a batch file I made a while back out of common-enough
image manipulation tools.  (Win7 machine)  which does an autotrace into
SVG, which I then took and made an animated GIF out of.

I know it's possible to do similar things in ImageJ and although I'm not
100% sure what you're looking for I know that getting a good segmentation
can be the trickiest part, based upon a good binary, on top of which other
determinations (measurements etc) rests.

convert -magnify %1 pgm: | mkbitmap -x -f 50 -t 0.48 | potrace -s -t 5 >
%1.svg

This is my windows batch file.  Programs are convert, mkbitmap and
potrace.  I don't remember where I got them all (convert is imagemagick,
that much I know) but I think they're common enough.

it produced this:
http://icopiedyou.com/wp-content/uploads/2020/01/turbs2.gif which I've done
in ImageJ before but there's always more than one way to do things. Don't
know if it's helpful or not but I enjoy trying things out. "Proof of
concept" more than "hard scientific" here.

Ken
[image: turbs2.gif]



On Mon, Jan 20, 2020 at 12:22 PM Herbie <[hidden email]> wrote:

> Dear Frank,
>
> attached (see the follwing link) please find an ImageJ-macro that does the
> proposed regional "Box Counting"-analyses for all images of a stack.
> fractalDim.ijm
> <http://imagej.1557.x6.nabble.com/file/t380516/fractalDim.ijm>
> (Perhaps you start with a single image, because the writing to the
> results-table takes a while...)
>
> Paste the macro code to an empty macro window (Plugins >> New >> Macro) and
> run it with either a single image or a stack open in ImageJ.
>
> Regional analyses are made for image selections that are 128pel wide.
> The increment is half the width, i.e. 64pel, which means an overlap.
>
> The filter parameters are not critical.
>
> The automatic "Otsu"-threshold scheme works quite well for the supplied
> image stack.
>
> The D-Column of the results table gives the fractal dimension as defined by
> "Box Counting"-approach. See "Examples" in:
> <https://en.wikipedia.org/wiki/Fractal_dimension>
>
> Please also see the ImageJ user guide sub-subsection "30.14.2 Fractal Box
> Count...":
> <https://imagej.nih.gov/ij/docs/guide/146-30.html#toc-Subsection-30.14>
>
> HTH
>
> Herbie
>
> :::::::::::::::::::::::::::::::::::::::::::
>
> Frank Shaffer wrote
> > Herbie,
> > The feature recognition only needs to work on one image. Then it can be
> > repeated for the other images.
> > If you find something that works, perhaps you can record it in a macro
> and
> > post it.
> > Thanks
> > Frank
> >
> > On Mon, Jan 20, 2020, 10:02 AM Herbie &lt;
>
> > l16@
>
> > &gt; wrote:
> >
> >> Dear Frank,
> >>
> >> initially you provided a single image and now you provide an image
> >> series.
> >> I'm totally unsure what you like to see in the end.
> >>
> >> With respect to the single image you wrote:
> >> "We would like to characterize the morphology (sizes, length scales,
> >> etc.)."
> >>
> >> Is this still what you want or what does the image sequence mean?
> >>
> >> If you like, I can provide the result of my initial idea which works for
> >> single images.
> >>
> >> BTW I've provided a link to the Wiki dealing with "Box Counting"-
> >>
> >> Regards
> >>
> >> Herbie
> >>
> >> :::::::::::::::::::::::::::::::::::::::::::
> >>
> >> Frank Shaffer wrote
> >> > I don't have experience with the fractal recognition technique.
> >> > If anyone would like to try their suggestions on an example video, you
> >> can
> >> > download it here:
> >> >
> >> > www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
> >> >
> >> > It's only 31 frames.
> >> >
> >> > If you technique works well, I will be happy to give you credit in our
> >> > publications.
> >> >
> >> > Thanks,
> >> > Frank Shaffer
> >> > FDShaffer.Net
> >> >
> >> > --
> >> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >>
> >>
> >>
> >>
> >>
> >> --
> >> Sent from: http://imagej.1557.x6.nabble.com/
> >>
> >> --
> >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >>
> >
> > --
> > ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>
>
>
>
>
> --
> Sent from: http://imagej.1557.x6.nabble.com/
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Frank Shaffer]

Professor Settles,
We had some discussions in the early 90's about RFI noise from CuV lasers. I've followed your work since.

Professor Savas is one of the great experimental fluid dynamicists of our generation to come out of Cal Tech.
He's still very active. But I doubt he's looked at his website in ten years.

I'll respond further to your PSU email.

Best regards,
Frank

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Peter Haub]

@Herbie
Hi Herbie, maybe you can explain how I can upload a file to the mail
server and attache the link in my email .. as you did in one of your emails.

(Is there a size limit ?)

Thanks
and regards
Peter


-------- Forwarded Message --------
Subject: Re: Turbulent Jet Morphology?
Date: Tue, 21 Jan 2020 19:25:24 +0100
From: Peter Haub <[hidden email]>
To: ImageJ Interest Group <[hidden email]>, [hidden email]



Hi Frank,

have you tried to analyze the optical flow in your video?

The the attached example (in the zip archive) - calculated with the FIJI
plugin 'OpticalFlow'.

Regards,
Peter


On 20.01.2020 14:08, Frank Shaffer wrote:

> I don't have experience with the fractal recognition technique.
> If anyone would like to try their suggestions on an example video, you
> can download it here:
>
> www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
>
> It's only 31 frames.
>
> If you technique works well, I will be happy to give you credit in our
> publications.
>
> Thanks,
> Frank Shaffer
> FDShaffer.Net
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Herbie]

Dear Peter,

I did the upload via Nabble:
<http://imagej.1557.x6.nabble.com/>

I'm not aware of a certain size limit, although there must be some
limitation.

=========================

Apart from this formal question, I should like to again point out that
the OP (Frank Shaffer) likes to characterize the *morphology* of jets in
*single* images. Originally he wrote:

"Attached is an image from a high-speed video showing turbulence eddies
of a submerged jet (UC Berkeley Fluid Dynamics Lab).  We would like to
characterize the morphology (sizes, length scales, etc.)."

More recently Frank confirmed that he is interested in jet
characterization from single images:

"The feature recognition only needs to work on one image. Then it can be
repeated for the other images."

That said, I'm not sure if "Optical Flow" that is based on image
sequences, not single images, may help to characterize the jet morphology.

Regards

Herbie

::::::::::::::::::::::::::::::::::::::::
Am 22.01.20 um 12:04 schrieb Peter Haub:

> @Herbie
> Hi Herbie, maybe you can explain how I can upload a file to the mail
> server and attache the link in my email .. as you did in one of your
> emails.
>
> (Is there a size limit ?)
>
> Thanks
> and regards
> Peter
>
>
> -------- Forwarded Message --------
> Subject:     Re: Turbulent Jet Morphology?
> Date:     Tue, 21 Jan 2020 19:25:24 +0100
> From:     Peter Haub <[hidden email]>
> To:     ImageJ Interest Group <[hidden email]>, [hidden email]
>
>
>
> Hi Frank,
>
> have you tried to analyze the optical flow in your video?
>
> The the attached example (in the zip archive) - calculated with the FIJI
> plugin 'OpticalFlow'.
>
> Regards,
> Peter
>
>
> On 20.01.2020 14:08, Frank Shaffer wrote:
>> I don't have experience with the fractal recognition technique.
>> If anyone would like to try their suggestions on an example video, you
>> can download it here:
>>
>> www.FDShaffer.net/Berkeley/Test_009_Fluorescent_enhanced.avi
>>
>> It's only 31 frames.
>>
>> If you technique works well, I will be happy to give you credit in our
>> publications.
>>
>> Thanks,
>> Frank Shaffer
>> FDShaffer.Net
>>
>> --
>> 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