Moving to smarter document management

NEW YORK: Invention, innovation and integration are the key research drivers at Xerox Corporation, a $15.7 billion document management company. The computer mouse, graphic user interface and ethernet are some of its important contributions to the world.

Last year, the company demonstrated the voice-activated technology called Clarissa, developed at its European Research Centre. Clarissa is now being used on the International Space Station.

On the research initiatives at Xerox, Sophie Vandebroek, chief technology officer and president, Xerox Innovation Group, said, “Xerox has had a strong history of innovation. The basic research that we do as a company is truly focused on systems, solutions, services, smart material among others. We have about 5,000 world class engineers and scientists over the world helping us carry out path breaking research and have over 50,000 global patents to our name.”

Some interesting next generation technologies are currently taking shape in its labs while a few of them are ready to make their debut.

Intelligent Redaction

Xerox scientists are working on a technology called intelligent redaction to help companies deal with managing document content access. Currently there are many issues that make it necessary to protect particular information within complex documents. For instance, in the medical field, it is desirable that the sensitive medical data be seen only by those who really need to see them. Similarly litigation practitioners, loan lending institutions, pharmaceutical companies and government agencies face similar problems. Yet another issue causing problems for companies managing their document content is the sheer numbers of documents to be tracked and secured.

According to Dr Shriram Revankar, Xerox fellow and manager, “Presently organizations facing this issue are using ad hoc approaches that are neither secure nor consistent. That is because there are no existing tools that provide sufficient content analysis and security, so there are not adequate systems available today to help companies protect sensitive data embedded within documents.”

He said, “Intelligent redaction will allow document owners to more easily control who sees what in a document and who has access to particular documents and information. It also creates a behind-the-scenes audit trail that could come in handy should the document or information be compromised.”

Now you see it; now you don’t

Redaction is the ability to control what someone sees. For example, a censored document. Traditionally this has required intensive manual processing of the document to identify sections to censor, leading to cumbersome and unpredictable management of different versions of the same document for different audiences.

Intelligent redaction takes this concept one step further by allowing the document itself to hide or expose information or data within it based on who is accessing the document. The document appears different to different people because intelligent reaction software ensures that the document itself can control who has permission to see varying subsections of it. Particular levels of clearance are assigned to certain sections of the document so that when a person accesses that section they can automatically see only the sections they are permitted to see.

Intelligent redaction displays or hides restricted sections of the document and the person reading the document may not even know that the sections are missing.

How does it work?

This is an important problem to solve, but a variety of underlying technologies are at play here to identify and protect those parts of the document that are sensitive given the current audience of the document, involving sophisticated natural language processing and interaction with enterprise rights management servers, as well as the document editor that is being used. For example, if the document was created using MS word, intelligent redaction would take the form of a software plug-in for MS Word that would identify redactable content, and automatically interact with the rights management server to ensure the intelligent redaction happens seamlessly.

Responding to queries on the commercialization of this technology, Anne Mulcahy, CEO and president, Xerox, said, “I would hope that the entry into the market is not too far from now. It is an interesting piece of technology and obviously it is a big piece of what we offer in our litigation service practice. But the one great thing about having a services business is that it reduces the time to market from the actual research centre into the solutions that we bring to our customers. In fact, some of that is being played out to our customers now. And as we get more experience that we build into the platform so that it is available on a broad basis soon.”

Specialty imaging

Some other interesting technologies being fine-tuned at Xerox are in the area of Specialty Imaging. Not too long ago, high-quality paper, an official seal, colored images, engraving or a watermark were enough to convince recipients that a document was authentic. For centuries it worked for stock and bonds, currency, birth certificates, identification papers, insurance policies and other “official’ papers. But no longer! The widespread availability of scanners, image-editing software and high-quality digital printers enables just about anyone to turn out official looking documents that once required a skilled offset press operator or an engraver. So how can those who issue the papers and those who receive them be certain that they are dealing with authentic copies and not forgeries?

Researchers at Xerox have been working on this problem. And they have discovered that while digital printing is part of the problem, it can also be part of the solution. Some of their innovations take advantage of the variability enabled by digital printing technology; others also involve the unique properties of the toners or dry inks that are used in xerographic printing. These will increase the tools available to those issuing valuable documents and thus aid in guarding against counterfeiting.

“Run-lengths of one, made possible by digital printing, are changing the economics of counterfeiting,” said Reiner Eschbach, a Xerox research fellow. “In the world of static information, once you have forged the basic form, you can replicate it again and again, reaping big gains. But with digital printing, each original can be different; you do not have a universal key after counterfeiting just one, and that reduces the incentive.”

Eschbach said that new technologies will make it possible to add a layer of security in a relatively inexpensive way on birth certificates, transcripts, insurance policies, titles, stock certificates, and other applications where it really was not possible to do so before. Eschbach calls the Xerox innovations “Specialty Imaging,” which refers to the underlying digital imaging technologies involved.

More innovations

Microtext Specialty Imaging Fonts print text so small that one needs a magnifying glass to read the words. It is smaller than 1 point—less than 1/70th of an inch or 350 micron tall. Issuers already hide microscopic words in the design of credit cards, checks and currency as a deterrent to counterfeiting. With the innovation at Xerox, microprinting can be carried to the next level as it can make important documents more secure by individualizing the micro printed letters and numbers.

GlossMark Specialty Imaging Fonts and images are produced with Xerox software and provide different levels of gloss on a page in a predetermined manner. This new printing technology can be used to embed hologram-like words and images, which can be seen as an additional and separate image when the paper is tilted, but cannot be scanned or copied on conventional devices. GlossMark technology has two major advantages over holograms and other laminates that are used to create similar effects: the GlossMark process does not require any additional printing steps or incremental costs and it accommodates variable information, like a name, a time stamp or a code.

CorrelationMark Font is being developed from a collection of Xerox technologies that use correlation and interference. Two correlated halftone patterns are printed together with some spatial displacement between them. To the naked eye, CorrelationMarks look like a normal image. However, a simple assisting device, such as a plastic grate placed over the image, reveals a secret message in the patterns.

UVMark is another way to add a layer of security to an ordinary printing job. With Xerox’s innovation, a digital printer can create variable text that is hidden until exposed to ultraviolet light. The method is unique because it is accomplished without the use of special ultraviolet inks or toners. Instead it takes advantage of the fluorescence that paper produces have already formulated into paper to enhance its “whiteness.” Xerox scientists discovered that by printing with the right combination and amount of existing toners, they could vary how much of the paper’s fluorescence was absorbed by the toner. This discovery is still under development in the Xerox labs.

Automatic Image Enhancement

Another technology developed by Xerox scientists at its Europe Centre is the Automatic Image Enhancement (AIE). While software exists to manually "fix" digital image problems, but it requires time and a skilled craftsman. The AIE automates color correction in Xerox systems. The basic AIE algorithms brightening underexposed images, sharpening fuzzy prints, or burning the haze off a vacation scene to let the bright colors shine through. It can correct the shortcomings found in images clipped from a Web page or compressed too tightly for a faster e-mail.

According to Monica Beltrametti, VP and director, Xerox Research Centre Europe “Automatic Image enhancement combined with our image categorization technology we can perform "class based" image enhancement. It recognizes image content to control enhancement parameter for example make flowers more colorful, give snow a high exposure. While some other image categorization applications include personal photo management; enable search and categorization of images as for text; web image search; corporate/ stock image and video repositories; and human computer interaction.”

© CyberMedia News