Quote:
Originally posted by Kimmykim
But if there are a whole bunch of images out there named pic1, pic2, pic3 and they are all the same size for instance, then how would your system discern between all the pic1s to know what was what?
For every person that renames the same picture differently you are going to have to review the same picture again and again in order to continue to catalog it?
And this is before they are resized, recompressed etc?
Be simple in the response here, I'm a technical dunce...
|
In simple terms, "Yes". If the image is cropped or is resized it will be given a different MD5 string [MD5 strings are 32 characters], if Sniffy does not recognize the MD5, it treats the image as a new image and it needs validated.
Our database is not of the images, but instead the hash strings.
So the image is not really a duplicate at the bit and byte level, just to humans who reason the similiarities :-)
Side note: If image is renamed, the md5 is still the same. Only by changing the bits of the file, not the filename, will the md5 value be different