high speed video with varying illumination

[Franklin Shaffer-2]

I'm taking high speed videos of particles at 10,000 frames per second, but the light source I'm using for illumination is varying in brightness at 60 hz.  
The result is that the brightness of each image is different, even when the object being viewed is not changing.

Anyone have any suggestions for image processing steps to correct for the varying illumination?

(I don't have the option of getting a different light source.)

Thank you,

Frank Shaffer
Research Engineer
USDOE National Energy Technology Laboratory
Computational Science Division

[Philip Ershler]

On Oct 16, 2008, at 4:21 PM, Franklin Shaffer wrote:

> I'm taking high speed videos of particles at 10,000 frames per  
> second, but the light source I'm using for illumination is varying  
> in brightness at 60 hz.
> The result is that the brightness of each image is different, even  
> when the object being viewed is not changing.
>
> Anyone have any suggestions for image processing steps to correct  
> for the varying illumination?
>
> (I don't have the option of getting a different light source.)
>
> Thank you,
>
> Frank Shaffer
> Research Engineer
> USDOE National Energy Technology Laboratory
> Computational Science Division

Is there any possibility that the light source could be driven off of  
a well regulated DC power supply?


Phil

[Raymond Lillard]

Franklin,

Most times the best solution is the simplest.
Get a DC power light source and save yourself
a lot of work.

The hard way is to load the images into a stack,
reslice the stack such the the 60 Hz is displayed
in each frame.  The 60 Hz will appear as light and
dark bars (either vertical or horizontal, depending
on how you rotated the stack).  Filter out the bars
(probably FFT) and then rotate the stack back to its
original orientation.

Ray




Franklin Shaffer wrote:

> I'm taking high speed videos of particles at 10,000 frames per second, but the light source I'm using for illumination is varying in brightness at 60 hz.  
> The result is that the brightness of each image is different, even when the object being viewed is not changing.
>
> Anyone have any suggestions for image processing steps to correct for the varying illumination?
>
> (I don't have the option of getting a different light source.)
>
> Thank you,
>
> Frank Shaffer
> Research Engineer
> USDOE National Energy Technology Laboratory
> Computational Science Division
>

[Justin Walker-3]

If you don't mind them being reduced to 2-bit images, a simple automatic
thresholding should take care of it.

However, you should really look into an 80 kHz light (I, too, use a 10,000
fps camera to track particles, and the 80kHz light makes a huge difference).

- Justin Walker
University of Maryland

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Franklin Shaffer
Sent: October 16, 2008 6:21 PM
To: [hidden email]
Subject: [IMAGEJ] high speed video with varying illumination

I'm taking high speed videos of particles at 10,000 frames per second, but
the light source I'm using for illumination is varying in brightness at 60
hz.  
The result is that the brightness of each image is different, even when the
object being viewed is not changing.

Anyone have any suggestions for image processing steps to correct for the
varying illumination?

(I don't have the option of getting a different light source.)

Thank you,

Frank Shaffer
Research Engineer
USDOE National Energy Technology Laboratory
Computational Science Division

[Frederic V. Hessman]

If the particles stay in the field of view or if the number of  
particles is large and hence approximately constant, then try  
subtracting the background (presumedly also varying due to scattering)  
and then normalize by the integrated brightness of the whole image.

Rick

------------------------------------------------------------------------------------------------
Dr. Frederic V. Hessman     [hidden email]
Institut für Astrophysik          Tel.  [phone]
Friedrich-Hund-Platz 1         Fax [phone]
37077 Goettingen                 Room F04-133
http://www.Astro.physik.Uni-Goettingen.de/~hessman
-------------------------------------------------------------------------------------------------
MONET: a MOnitoring NEtwork of Telescopes
http://monet.Uni-Goettingen.de

On 17 Oct 2008, at 12:21 am, Franklin Shaffer wrote:

> I'm taking high speed videos of particles at 10,000 frames per  
> second, but the light source I'm using for illumination is varying  
> in brightness at 60 hz.
> The result is that the brightness of each image is different, even  
> when the object being viewed is not changing.
>
> Anyone have any suggestions for image processing steps to correct  
> for the varying illumination?
>
> (I don't have the option of getting a different light source.)
>
> Thank you,
>
> Frank Shaffer
> Research Engineer
> USDOE National Energy Technology Laboratory
> Computational Science Division

[nckrzan]

Hello,

I work with similar (maybe no so advanced camera, but similar)
Speedcam Weinberger - Macrovis 1000 fps - with 1200x1000 pixel resolution

In our research we use various kinds of halogen illuminations - and
observe no influence of the brightness connected with the AC frequency 60
hz

Try our solutions:

Hedler-HT19s-e11  - 3 x 650W  halogen (it is common use in the
professional photography)

or halogen with the fibre wire made by Schott North America Inc. KL 1500
LCD or KL 2500 LCD

Best regards,

Dr. Marcel Krzan
"Dispersed System" research group
Institute of Catalysis and Surface Chemistry
Polish Academy of Sciences
ul. niezapominajek 8
30-239 Cracow
Poland

[Franklin Shaffer-2]

I thank all who have responded.  I am looking for an image processing solution to this problem.  
I have a stack of more than 10,000 frames of particles.  The brightness (grey level) of the particles varies significantly from frame to frame, not only because of the illumination source.  
My goal is to detect and locate the centroid of most of the particles in each frame.  Here are the steps I'm using in ImageJ
1. Adjust Brightness/Contrast
2. Otsu Thresholding
3. Adjust Threshold
4. Analyze Particles

This misses most of the particles in dim images.  If anyone has any suggestions to deal with the problem of varying brightness (grey level) from frame to frame, I would appreciate your advice.
Thank you,
Frank Shaffer
USDOE NETL


>>> "Marcel Krzan PhD" <[hidden email]> 10/17/2008 3:32 AM >>>
Hello,

I work with similar (maybe no so advanced camera, but similar)
Speedcam Weinberger - Macrovis 1000 fps - with 1200x1000 pixel resolution

In our research we use various kinds of halogen illuminations - and
observe no influence of the brightness connected with the AC frequency 60
hz

Try our solutions:

Hedler-HT19s-e11  - 3 x 650W  halogen (it is common use in the
professional photography)

or halogen with the fibre wire made by Schott North America Inc. KL 1500
LCD or KL 2500 LCD

Best regards,

Dr. Marcel Krzan
"Dispersed System" research group
Institute of Catalysis and Surface Chemistry
Polish Academy of Sciences
ul. niezapominajek 8
30-239 Cracow
Poland

[Robert Dougherty]

Frank,

> My goal is to detect and locate the centroid of most of the particles in
> each frame.  Here are the steps I'm using in ImageJ
> 1. Adjust Brightness/Contrast
> 2. Otsu Thresholding
> 3. Adjust Threshold
> 4. Analyze Particles
>
> This misses most of the particles in dim images.  If anyone has any
> suggestions to deal with the problem of varying brightness (grey level)
> from frame to frame, I would appreciate your advice.

I'm guessing the dim images are noisy.  Perhaps apply a median filter as
step 0.5 or 1.5?  This would introduce an element of selectivity based on
spatial coherence, which Otsu thresholding seems to lack.

Bob
Robert P. Dougherty, Ph.D.
President, OptiNav, Inc.
Phone [phone]
Fax [phone]
www.optinav.com