I suggest actually timing some of this stuff before writing off the idea. If an image can be processed in 1 second, and you can parallelize it to 4 cores, some quick and dirty (and probably useless) math says 2 million images could be done in just under 6 days. I don't think that's too shabby for 2 million images.
Yes, definitely planning to parallelize but we're likely limited to four simultaneous instances so that will still take ~20-30 days.
Also, converting the (hi-res) images to JPEG before further processing may make the data somewhat easier to deal with. 25-30MB per image * 2 million would be 50-60 terabytes of data? Plus, how often would someone really benefit from having access to a 25-30MB TIFF vs a much smaller JPEG version in the same resolution? That alone could be reason to start with JPEG compression first.
Also, if you can parallelize it, it won't take 115 days.