More methods to combine different source detection results exists

[kenfus]

The current workflow is that two sources are the same if they are closer than X pixels to each other. In my opinion, this has clear cases where this does not work (e.g. for very bright sources or if the sources vary a lot in brightness).

With the total flux, gaussian fit, rotation of the gaussian fit and the overlap of those (e.g. if the gaussian fit of one sources overlaps "at least 1*σ into another source", then they are actually the same source). Probably, we can also use parameters given by the user to pyBDSF to calculate this.