We have a database with over 1,000,000 SHA1 hashes of copyrighted pictures. (we have no access to the real images though)
We need to scan a lot of workstations to see if they have these images illegaly copied onto their hard drives.
No problem of course if all these images are unmodified, we just hash the local images and if they are the same as in the database we have a "positive".
But if someone has altered the image a bit on his PC, it will not be recognised by the system, because the hash string will be different.
So we need a "similar to" or "partial positive" comparison. Vey much like the technology used in [login to view URL]
Can someobe build this for us?
Preferably in Ruby, but anything else is fine as long as it runs on Linux.
I could create a Python or C application (your choice) that uses OpenCV to detect images that have undergone resizing, color modification, or other simple transformations you may anticipate users performing on the images.
I would be happy to take on this project for you.
Hi, I am Harsh from Harrier Information Systems Pvt. Ltd. Harrier is a ISO 9001:2000 certified and also a Microsoft certified partner company. Please check PMB for more information about Harrier. Thanks, Harsh