Hello all. I am planning to build the microscope with the Delta Stage. I noticed that in the paper (Optica Publishing Group) it is said that Fourier ptychography is possible through the use of extensions. I am wondering if anyone has attempted this, and if successful could share some images?
Thank you!
I don’t know wheter it has been done, but it is something that I do want to try.
I think that the ordinary OpenFlexure Microscope is probably the better platform for it. You should only need xy motion.
You also require the LED grid based illumination, or is this not unique only to the Delta Stage?
You can use the LED grid on either type of body, the illumination mounting geometry is the same. The LED grid was developed on the Delta Stage, and that is where the parts and instructions are.
Ptychography would normally use an aperture and move the sample, I thought.
Thanks for the information! That is good to know. I am still nearly learning about the Openflexure design.
Regular (real space) ptychography does indeed move the sample, Fourier ptychography instead moves the illumination angle.
Hi there!
I’ve previously spoken to someone at Heidelberg who had a go at Fourier ptychography - he used the Cartesian stage with an extra translation stage bolted on for illumination upright-style, but had some trouble with suspending his LED matrix with good alignment. He did get some decent images out; as far as I can tell he didn’t want to post to the forum yet. As far as I’m aware it hasn’t been tried in a Delta stage!
Interesting! I’m interested in how exactly the extra translation stage was used!
Do you happen to know what was his achieved illumination NA and effective NA were? I can imagine that LED matrix alignment would certainly be tricky. I think there are ways to correct for this right? I think it could be possible to infer the true position of each LED in space.
.
I am planning to work on this topic for my mphil thesis( OFM + Ptychography + Malaria)
Hi, we talked at FOSDEM last year!
I’ve not been able to work on this for a while and sadly never got to the point of successfully using Fourier Ptychography at all. I tried it on two different low-cost setups mostly with the USAF 1951 resolution target which is used in many papers about this method. I got a few reconstructions with improved resolution but never clean looking reconstructions. Always with lots of artifacts from the reconstructions and if I remember correctly, reconstructions from 5 images (central LED + 4 neighboring ones) worked consistently but not with more of them.
The “correct” approach would probably be to first replicate Fourier Ptychography on a modified high-end microscope and then moving down to low-cost components. But I didn’t have access to one and couldn’t narrow down my troubleshooting.
I think the biggest limitation of the setup in the images I attached is getting the LED matrix and the camera+objective stack to be parallel/horizontal and to find a good workflow to center the LED matrix (to pick one LED that is right on the optical axis). In our makerspace group in Heidelberg we are currently building kinematic mounts for an optical breadboard, this should allow for a more robust alignment than what I currently have.
Also, what is not really visible here, I modified the optical tube of the open flexure microscope to be 160mm long and not include that additional lens because I thought that the demagnification caused by it might lead into undersampling.
