Given be a set of signals (e.g. images) coming from one or more sources (e.g. different types of cameras) is it possible to find out, which signals come from which source?

I consider this question very relevant for the pre processing step of data, for example medical data. Assuming a data set was generated by only one machine is very naive and may ultimately used in the pratice.

Therefore I am looking for material to work through, that would allow me to tackle this problem, for example by finding an above described clustering algorithm.

Thus far I found a method identifiying camera models, namly here: K-unknown models detection through clustering in blind source camera identification.

However, I am more interested in the medical use cases of sources like x-ray, ultrasound etc., thus I would be very thankful for someone pointing me into the right direction.

  • $\begingroup$ Just to clarify my understanding, are you attempting to perform clustering or classification? $\endgroup$ – Engineer Nov 15 '19 at 19:46
  • $\begingroup$ I am interested in clustering, where the number of cluster is unknown. The sources (for example cameras) might be similar, however still have some finger print by which they can be recognized. Of course, if one knows the fingerprint, one can do classification, however it is enough for me at the moment to simply know, if two signals come from two different sources. $\endgroup$ – Imago Nov 15 '19 at 19:54
  • $\begingroup$ Parse the DICOM file? $\endgroup$ – user28715 Nov 15 '19 at 22:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.