Xerox unveils smart-document software

The lives of office file clerks may get a little easier if Xerox delivers on its promise of new smart-paper technology.

Researchers from the company's Palo Alto Research Center (PARC) on Thursday announced three new technologies that augment digital and paper documents.

First is the hybrid categorizer, software that uses computer vision and other machine learning techniques to understand text and general images on a digital page. Then it links the text and images so the document can be categorized. For example, an insurance company could use the software to digitally file a document containing an image of a car into an auto claims folder, vs. one for home claims. The software would automatically complete the task without the help of a file clerk.

The second in Xerox's software cadre deals with mining pages for sensitive information. The technology, called "intelligent redaction," is a content detection tool that can encrypt sensitive sections or paragraphs of a digital document, or it can automatically remove confidential data from the page. For example, if a hospital needs to send a patient's medical folder to another doctor as part of a legal case, the technology could encrypt the information about the patient that needs to remain private.

Continuing on the spy movie theme, Xerox also announced a new security feature for paper documents. Its so-called infrared specialty printing technology lets customers create and embed a hologram onto a physical piece of paper to prevent copying. An infrared light source can authenticate the document by revealing its hidden text. Xerox CTO Sophie Vandebroek said the technology could be used for birth certificates, store coupons and movie tickets.

"There's an amazing opportunity to leverage these technology across many markets," Vandebroek said.