| layout | title | date | categories | header | author | author_profile | ||||
|---|---|---|---|---|---|---|---|---|---|---|
single |
DeconvOptim.jl |
2022-02-17 18:16:06 +0100 |
|
|
Felix Wechsler |
true |
Recently, we worked on a deconvolution toolbox written in Julia Lang, called DeconvOptim.jl.
It can deconvolve multi color data very efficiently with a quality equal to Huygens Deconvolution.
The routine is very generic and based on a loss function approach which is minimized with the automatic differentiation Zygote.jl and the optimizer package Optim.jl.
Also we put a lot of emphasize on the performance of DeconvOptim.jl, as shown here.
Using Julia's CUDA ability we could outperform any existing deconvolution package but maintain the same quality measure - here under the normalized cross correlation (NCC) value.
See also that video
<iframe width="560" height="315" src="https://www.youtube.com/embed/FodpnOhccis" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>