I have used DANPOS in the past and I'm quite happy with it. It is reasonably fast and outputs both peak tables and bigwig files. I also enjoyed reading the original DANPOS paper.
Unfortunately DANPOS relies on an extremely old samtools version, while fine for quick manual peak calling, makes it unsuitable for integration into a pipeline meant for multiple different systems. It's just another dependency i'd have to add.
What I did was to install the old samtools version without adding it to my PATH so there was no interference with the newer samtools version already installed. Then I modified the python code to replace the samtools commands by the absolute path the old samtools. It was not too much trouble in my case.
See here, it lists and compares many nucleosome callers http://bib.oxfordjournals.org/content/17/5/745
What an excellent resource thank you.