You can use these routines to deal with the files written by the ThermoNoran "InterVision" software which runs on an SGI. My impression is that these may just be some sort of special SGI image stack format.
IDL> a = read_noran('stack.noran')
Note that these image stacks have a curious feature: the last few images stored in the file belong to the beginning of the data set. So, read_noran by default swaps the very last image to the front of the file. For small files this is sufficient, say, files with 10 images or fewer. For larger files, it will be more than one image that is at the end of the file, and so this won't help very much. In fact, it could be a problem because it would take up more memory as it does the swapping. So, you can use the keyword /lomem to skip this action in read_noran and thus save on memory.
I've written some code in epretrack that fixes this problem, more information coming soon.