Mohd, more information is needed to answer your question. Just to be clear, I assume that you are referring to a reconstructed wavefront. The second clarification is in regard to "extended sources", are you referring to sources that span multiple isoplanatic angles? At the most basic level, image reconstruction is achieved by applying the conjugate phase (from the reconstructed wavefront) to the distorted image phase. Not sure if that helps, but it is a start.

Hi thanks for your reply Jeremy .

My team is working to reconstruct solar images from obtained reconstruct wavefront !!!

In matlab form , we obtained this wavefront by introducing turbalance , finding centroid using cross correraltion , using bling deconvolution , and fitt the zernike upto 21 ---while low order abberation coming more----we are foucs on solar image recosntruction !!!

I really thankful to you if you can give some valauble inputs in this regards :

1.How i can start it

2.Which algorithm i can used in this regard

3.Which data reduction platform more favourable for it

4.How i can get reconstructive soalr image by mimimizing the turbalance effect as much as possible to get the best image in real time mode.

Experimentally , we are doing this by adaptive optics kit to reconctruct the wavefront in which we have the facility of SHWFS and Deformable mirror !!!

Thanks :

Mohd Faisal Nawaz

Mr Jeremy : Can this paper sort out this issue , as u r telling to focus on phase conjugate things !!!

http://www.google.co.in/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&ved=0CCYQFjAB&url=http%3A%2F%2Fmenehune.opt.wfu.edu%2FMenehune%2Fplemmons%2Fmpdr.ps.gz&ei=SMh-UMmUGcnJrAffmoHwBw&usg=AFQjCNFoSaIqdzlcvl8CXZI53PAq5fdmtg&sig2=ioTlZki66vz1YSfPyaLOJA

Mohd,

The paper you reference uses a phase diversity approach identify the conjugate phase. It is my understanding that phase diversity techniques use a multiple frames blurred by turbulence and additional frames subject to turbulence blur plus known defocus to estimate the phase distortion and the object via non-linear optimization. Are you proposing to use phase diversity as a solution? Above you seem to be inquiring about a numerical method suitable for recovering the wavefront phase. The methods mentioned in the paper below are well-suited to this task. If you plan on using phase-diversity then the method outline in the paper should work well. If you do not plan on using phase diversity I believe you will need to identify a different likelihood function, but the numerical approach will still be appropriate. Do some research on Multi-Frame Deconvolution Methods(MFBD) in atmospheric imaging.

I also understand that you are looking to perform reconstructions in real-time. Real-time reconstruction using MFBD is unlikely. Given sufficient computing power you can do real-time reconstruction using speckle-imaging methods. Woger, Mikurda, and von der Luhe wrote a program called KISIP that would likely work for this purpose. See below.

https://forge.kis.uni-freiburg.de/kisip

So, to answer each of your follow on questions:

1) Sounds like you have an answer here.

2) Same here, the algorithms mentioned in the paper will work just fine.

3) Not sure what you're asking here

4) If you are using MFBD you should be jointly estimating the distortion (phase) and the object. So, the object/image will be produced as a byproduct of the optimization.

Jeremy

Hi ,

I am Very thankful to your sir , on Your Valuable suggestion :

What we did to reconstruct wavefront :

Took the solar image , convolve with a Guassian function with a variance of 10 , and make a lenslet arrays of a pixel size 50 by 50 , later than we multiply phase screens with it to get the turbulated images , then with reference of image , we find the centroid data using cross correlation , x y slope of each , and then find out the 21 order zernike , through which we obtained the Reconstructive wavefront !!!

we do this work on Matlab Platform !!!!

I find in some papers in which Principle component analysis Method , Frozen Flow Hypothesis Method , MUlti Frame Blind Deconvolution Technique , Phase Diversity etc taken by researchers for restoration turbulated images !!!!

Now please suggest me , that how i can proceed to reconstruct the solar image with all such data !!!

--

With Regards :

Mohd Faisal Nawaz

If all you have is the reconstructed wavefront, there's not too much you can do. It is just an estimate of all the phase screens you put in the system. You can try to compute the resultant point spread function. Is that what you're trying to do?

Interesting discussion,

I think that if you have the estimated wavefront, you can use the technique "Deconvolution from wavefront sensing". Its a deconvolution process in which you compute the restoring filter using the information of the estimated wavefront.

I give you one reference, but please see the reference section, and specially the work by Primot et. al.

http://dx.doi.org/10.1364/OE.11.000761.

I suggest you also the book of M. Roggeman, "Imaging thourgh turbulence"