According to a filing issued to the U.S. Patent and Trademark Office, patent 7,158,961, Google is working on deploying a "similarity-engine" that would compare documents and websites for redundancy. Most searches online produce similar results or exactly the same results. By eliminating the dupes, Google should be able to provide the user with the most relevant information.
The similarity-engine will be based on creating and calculating differences and sums in vectors. That means comparing differences in text and images would be plausible. Google also hinted that the implementation could be used to compare common office productivity documents. Don't get too excited though. Google's similarity-engine project is not unique: over 15 patents for similarity-engines have been filed over the last 10 years.
News source: DailyTech