Hi everyone.
I am working with developing inverse solver of EIT based on Artificial Neural network.
What I want to do is finding inverse solver that can reconstruct arbitrary shape of conductivity input.
In order to do that, input conductivity is totally random : position and number of elements that conductivity is changed are random.
I have been studying this topic because ,at least in my many researches, most of research seems to deal with using ANN to input of an already given shape such as a circle.
But I think many experts who are much better than me have thought of me like that. It would be great if I could hear the advice from someone who tried this part. Or, if you know the case of research that has been done on this topic, can you please let me know?
Thank you for your attention.
Best regards
Look at these fresh items(2018):
- http://www.ein.org.pl/sites/default/files/2018-03-11.pdf
- https://ieeexplore.ieee.org/abstract/document/8388351/
Typically the iterative solvers uses the background conductivity instead of an arbitrary conductivity or shape.
- Badges
- Science topic
- Similar topics
- Philosophy
- Ethics