Well the SHA-2 digest returned by this function IS different from what you get when you hash the file, because this funtions hashesh the pixels themselves.
This is very useful when you need to identify identical images, because if you just hash the files, you may get different hashes even if the image is the same pixel by pixel because of differences in metadata or different format (ie PNG vs BMP) or perhaps stray bytes at the end of file etc...
There is no point in posting code that hashes files, that is completely irrelevant here.
Also, I doubt you really need to pick a different hashing algo, SHA-2 should be perfectly fine for this job.
