*****
To join or leave the confocal microscopy listserv or to change your email address, go to:
https://lists.umn.edu/cgi-bin/wa?SUBED1=confocalmicroscopy&A=1
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Emmanuel,

The description of your issue (/"the GFP channel image is about 1 to 2 
pixels larger than the RFP image in every corner"/) suggests that it 
origins from lateral chromatic aberration, which can be seen as a 
dependence of the magnification of your system with wavelength 
(emphasized in the corners), rather than from misalignment between your 
two cameras.

Either way, I agree with Ian: tuning the hardware perfectly is 
impossible. However, improvements can be obtained via software image 
processing. The correction of co-registration inaccuracies, either 
between two cameras or two colors, can be achieved using suitable 
quality control solutions.

Argolight provides such solutions. In particular, a fluorescence pattern 
called "field of rings" allows to measure the "lateral co-registration 
inaccuracy" of fluorescence microscopes within the entire field of view 
and provides linear transformation parameters to correct for it. 
Protocols and guidelines can be found in the related documentation: 
https://argolight.notion.site/Lateral-co-registration-accuracy-f20381d9231144d7b8e07827f3ac1de0

Hope this helps,
Best regards,
Arnaud

*Arnaud ROYON, Ph.D.*
Argolight
Cité de la Photonique, Bat. Elnath
11 avenue de Canteranne
33600 Pessac, FRANCE
Email: [log in to unmask]
Tel: (+33) 5 64 31 08 50
Web site: www.argolight.com <http://www.argolight.com>

Le 18/04/2024 à 15:44, Ian Dobbie a écrit :
> *****
> To join or leave the confocal microscopy listserv or to change your email address, go to:
> https://lists.umn.edu/cgi-bin/wa?SUBED1=confocalmicroscopy&A=1
> Post images onhttp://www.imgur.com  and include the link in your posting.
> *****
>
> Hi Emmanuel,
>
> The reality is that tuning the hardware perfectly is impossible. Effects at this level are objective dependent (ir “identical” objective might have magnification differences of 0.1% between colours (ie 2 pixels over 2,000). In fact if you do super resolution imaging and are locking at small distance measurement between colours you have to ensure you correct for non-linearities between colours, where the images might align perfectly at both edges but have shifts in other regions.
>
> With the reality that you cannot align perfectly in hardware you should always take calibration data and develop a workflow that includes a (3D) alignment step, or distance correction. To pay particular attention to Z shifts as these are often larger than xy shifts. Dispersion in your sample will linearly scale the image in Z with refractive index change. The oil/coverslip component should be mostly taken account of by the lens, but the sample component cant be solved in the objective. Any small distance measurement must also include a Z component to be realistic.
>
> Ian
>
>
>> On Apr 16, 2024, at 5:22 PM, Emmanuel Levy<[log in to unmask]>  wrote:
>>
>> *****
>> To join or leave the confocal microscopy listserv or to change your email address, go to:
>> https://lists.umn.edu/cgi-bin/wa?SUBED1=confocalmicroscopy&A=1
>> Post images onhttp://www.imgur.com  and include the link in your posting.
>> *****
>>
>> Dear Zdenek and Cătălin,
>>
>> Thank you for your replies! Indeed, the "distance" of the camera relative
>> to the camera-port lens sounds like an intuitive parameter to adjust.
>>
>> So the question is, have you adjusted that distance before and achieved a
>> 1-2 pixel zoom in/out in this manner? I assume that there should also be an
>> optimal distance for proper focus
>>
>> The picture I attached was imaged with beads. We do a lot of
>> co-localization experiments, so it would be good if we managed to tune the
>> hardware flawlessly.
>>
>> Thank you for sharing your experience.
>> Best wishes,
>>
>> Emmanuel
>>
>>
>> On Mon, 15 Apr 2024 at 17:31, Cătălin Pavel<[log in to unmask]>  wrote:
>>
>>> *****
>>> To join or leave the confocal microscopy listserv or to change your email
>>> address, go to:
>>> https://lists.umn.edu/cgi-bin/wa?SUBED1=confocalmicroscopy&A=1
>>> Post images onhttp://www.imgur.com  and include the link in your posting.
>>> *****
>>>
>>> Hi Emmanuel,
>>> It looks that one of the camera is closer to the microscope than the
>>> other. Try to check that all the connections are flush and tight.
>>>
>>> Catalin
>>>
>>>> On Apr 15, 2024, at 09:30, Emmanuel Levy<[log in to unmask]>  wrote:
>>>> *****
>>>> To join or leave the confocal microscopy listserv or to change your email address, go to:https://lists.umn.edu/cgi-bin/wa?SUBED1=confocalmicroscopy&A=1
>>>> Post images onhttp://www.imgur.com  and include the link in your posting.*****
>>>>
>>>> Dear All,
>>>>
>>>> We have a W1 spinning disk with dual cameras. We recently had them aligned, and upon close inspection, we see that the GFP channel image is about 1 to 2 pixels larger than the RFP image in every corner (we used a 60x objective, and the cameras are primeBSI express with 6.4um pixels, so the image is off by 100-200nm in every corner). you can download the two images as a composite here if you are curious:
>>>> https://drive.google.com/file/d/1TWIxwiZF6uf3FspyA96AX6r092PB_102/view?usp=sharing
>>>>
>>>> What would be the best way to solve this issue? Shouldn't the standard camera mounts enable us to correct this?
>>>>
>>>> Thanks for your help and comments,
>>>> Best wishes,
>>>>
>>>> Emmanuel