measuring uniform illumination

[Richard Murray-3]

Hi Everybody,

Is it possible to use Image J to measure whether or not your getting uniform illumination using a
point, line or field spread function? And if it is possible how would I go about doing this?

Thanks in advance,

Richard Murray

[John Oreopoulos]

I think the easiest way to do this is snap a photo a blank sample  
with your microscope that has been setup for Kohler illumination. It  
should just look like featureless white image (make sure you're not  
saturating your camera, turn down the exposure or light source  
voltage if necessary). Then, using ImageJ, draw a long line anywhere  
on this image and use the "plot profile" function which plots the  
pixel intensity vs. the distance along this line that you've drawn.  
If you have uniform illumination, then the variation in intensity  
along that line should be 1% or less of the average value.

I'm sure there might be other ways to check for uniformity, but I  
think this is the simplest method.


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



On 22-Mar-07, at 11:28 AM, Richard Murray wrote:

> Hi Everybody,
>
> Is it possible to use Image J to measure whether or not your  
> getting uniform illumination using a
> point, line or field spread function? And if it is possible how  
> would I go about doing this?
>
> Thanks in advance,
>
> Richard Murray

[Richard Murray-3]

Could I also use the surface plot to obtain the same result?
--
Richard Murray BSc.
Nanoscale Biophotonics Research Group,
NCBES, Orbsen Biulding,
NUI Galway
Galway,
Ireland
email: [hidden email]
phone:[phone]

On Thu, March 22, 2007 4:34 pm, John Oreopoulos said:

> I think the easiest way to do this is snap a photo a blank sample
> with your microscope that has been setup for Kohler illumination. It
> should just look like featureless white image (make sure you're not
> saturating your camera, turn down the exposure or light source
> voltage if necessary). Then, using ImageJ, draw a long line anywhere
> on this image and use the "plot profile" function which plots the
> pixel intensity vs. the distance along this line that you've drawn.
> If you have uniform illumination, then the variation in intensity
> along that line should be 1% or less of the average value.
>
> I'm sure there might be other ways to check for uniformity, but I
> think this is the simplest method.
>
>
> 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
>
>
>
> On 22-Mar-07, at 11:28 AM, Richard Murray wrote:
>
>> Hi Everybody,
>>
>> Is it possible to use Image J to measure whether or not your
>> getting uniform illumination using a
>> point, line or field spread function? And if it is possible how
>> would I go about doing this?
>>
>> Thanks in advance,
>>
>> Richard Murray
>

[Jakub Sikora]

There are several very nice review articles dealing with this issue:
{Zucker, Cytometry, 2001), he suggests the use of flurescent plastic
slides (Chroma sells these but others also), surface plot can be very
nicely visualized when you apply LUT with defined percentage change,
this can be done in ImageJ very easily, example of this is on the www in
my signature, open the system evaluation (performed on confocal).
Hope I helped
Jakub

Jakub Sikora M.D.
Institute of Inherited Metabolic Disorders, 1st Medical Faculty, Charles
University
Ke Karlovu 2, Praha 2
Czech Republic
tel: [phone]
mobile: [phone]
fax: [phone]
http://udmp.lf1.cuni.cz/web2/facility/microscope/microscopy.htm

-----Original Message-----
From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
Richard Murray
Sent: Friday, March 23, 2007 12:32 PM
To: [hidden email]
Subject: Re: measuring uniform illumination


Could I also use the surface plot to obtain the same result?
--
Richard Murray BSc.
Nanoscale Biophotonics Research Group,
NCBES, Orbsen Biulding,
NUI Galway
Galway,
Ireland
email: [hidden email]
phone:[phone]

On Thu, March 22, 2007 4:34 pm, John Oreopoulos said:
> I think the easiest way to do this is snap a photo a blank sample with

> your microscope that has been setup for Kohler illumination. It should

> just look like featureless white image (make sure you're not
> saturating your camera, turn down the exposure or light source voltage

> if necessary). Then, using ImageJ, draw a long line anywhere on this
> image and use the "plot profile" function which plots the pixel
> intensity vs. the distance along this line that you've drawn. If you
> have uniform illumination, then the variation in intensity along that
> line should be 1% or less of the average value.
>
> I'm sure there might be other ways to check for uniformity, but I
> think this is the simplest method.
>
>
> 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
>
>
>
> On 22-Mar-07, at 11:28 AM, Richard Murray wrote:
>
>> Hi Everybody,
>>
>> Is it possible to use Image J to measure whether or not your getting
>> uniform illumination using a point, line or field spread function?
>> And if it is possible how would I go about doing this?
>>
>> Thanks in advance,
>>
>> Richard Murray
>

[Richard Murray-3]

Yes, you have been very helpful, thank you very much
--
Richard Murray BSc.
Nanoscale Biophotonics Research Group,
NCBES, Orbsen Biulding,
NUI Galway
Galway,
Ireland
email: [hidden email]
phone:[phone]

On Fri, March 23, 2007 12:01 pm, MUDr. Jakub Sikora said:

> There are several very nice review articles dealing with this issue:
> {Zucker, Cytometry, 2001), he suggests the use of flurescent plastic
> slides (Chroma sells these but others also), surface plot can be very
> nicely visualized when you apply LUT with defined percentage change,
> this can be done in ImageJ very easily, example of this is on the www in
> my signature, open the system evaluation (performed on confocal).
> Hope I helped
> Jakub
>
> Jakub Sikora M.D.
> Institute of Inherited Metabolic Disorders, 1st Medical Faculty, Charles
> University
> Ke Karlovu 2, Praha 2
> Czech Republic
> tel: [phone]
> mobile: [phone]
> fax: [phone]
> http://udmp.lf1.cuni.cz/web2/facility/microscope/microscopy.htm
>
> -----Original Message-----
> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of
> Richard Murray
> Sent: Friday, March 23, 2007 12:32 PM
> To: [hidden email]
> Subject: Re: measuring uniform illumination
>
>
> Could I also use the surface plot to obtain the same result?
> --
> Richard Murray BSc.
> Nanoscale Biophotonics Research Group,
> NCBES, Orbsen Biulding,
> NUI Galway
> Galway,
> Ireland
> email: [hidden email]
> phone:[phone]
>
> On Thu, March 22, 2007 4:34 pm, John Oreopoulos said:
>> I think the easiest way to do this is snap a photo a blank sample with
>
>> your microscope that has been setup for Kohler illumination. It should
>
>> just look like featureless white image (make sure you're not
>> saturating your camera, turn down the exposure or light source voltage
>
>> if necessary). Then, using ImageJ, draw a long line anywhere on this
>> image and use the "plot profile" function which plots the pixel
>> intensity vs. the distance along this line that you've drawn. If you
>> have uniform illumination, then the variation in intensity along that
>> line should be 1% or less of the average value.
>>
>> I'm sure there might be other ways to check for uniformity, but I
>> think this is the simplest method.
>>
>>
>> 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
>>
>>
>>
>> On 22-Mar-07, at 11:28 AM, Richard Murray wrote:
>>
>>> Hi Everybody,
>>>
>>> Is it possible to use Image J to measure whether or not your getting
>>> uniform illumination using a point, line or field spread function?
>>> And if it is possible how would I go about doing this?
>>>
>>> Thanks in advance,
>>>
>>> Richard Murray
>>
>