MITK presently at 1.44 on Nasdaq...I would buy it:
this today:
(COMTEX) B: Mitek Systems Announces Collaboration With Science Applic
B: Mitek Systems Announces Collaboration With Science Applications
International to Develop New Technology
SAN DIEGO (April 14) BUSINESS WIRE -April 14, 1999--Mitek Systems Inc.
(Nasdaq:MITK) and Science Applications International Corp. (SAIC)
Wednesday announced that they will jointly explore the potential for
applying new technology in the processing of loosely structured
documents.
John Thornton, Mitek's chairman and president, said: "To date, tools
for automating data extraction from loosely structured documents have
been quite limited. In general, existing tools have required extensive
hand crafting of rules for each application, which requires specialized
training, is very labor-intensive and is prone to error.
"Moreover, there is little or no generalization capability, requiring
extensive maintenance operations to keep pace with changes in incoming
documents. Our collaboration with Science Applications will explore new
methods of reading loosely structured documents by combining Mitek's
superior intelligent character recognition engines with SAIC's ability
to locate and extract data from a wide variety of loosely structured
documents."
Automated Document Processing
Historically, automated document processing systems have focused on two
extremes of document type: completely structured and completely
unstructured. Forms, such as tax forms, census forms, etc., are the
primary examples of completely structured documents.
Automated extraction of data from such documents relies on the fact
that each data element of interest appears on each form in a known
horizontal and vertical (X-Y) position. Data elements are located based
on these known positions, and are read using optical character
recognition (OCR) technology. The data elements typically are then
entered into a structured database for further processing.
At the other extreme of document processing are unstructured documents.
Examples include letters, reports, newspapers and magazines. In
general, such documents do not present information in repeatable X-Y
positions.
In automated processing, such documents typically are treated as
consisting of blocks of text. OCR technology is used to convert the
text from such documents for entry into full-text databases.
Many business documents, however, fall in between these two extremes of
document type. Examples include invoices, bills of lading and insurance
notices. Such documents typically have some type of header region that
includes identifying information such as a company name or invoice
number.
Much of the data of interest, however, is contained in subsequent
sections that may have only a generalized layout. As an example, an
invoice may list a variable number of items. Individual entries may
vary in vertical extent based on the length of each product
description.
Items such as ship-to locations, totals and payment dates may appear in
highly variable positions. Documents with such characteristics have
been referred to as flex-forms, pseudo-forms and partially structured,
or loosely structured documents.
Science Applications International
Over the past few years, SAIC has carried out an intensive research and
development effort to apply modern machine learning techniques in
document processing. This effort has resulted in a dramatic advance in
the ability to locate and extract data from a wide variety of loosely
structured documents.
This technology can deal effectively with data that can exhibit large
horizontal and vertical displacements, including movement across page
boundaries. Most importantly, this approach does not require any
explicit programming or rule crafting on the part of the user. Systems
based on this technology learn by example.
A user simply displays a few images of each document type and draws
bounding boxes around items of interest. The system automatically
learns how to locate the data. The algorithms employed possess very
powerful generalization capabilities, allowing many changes in document
layout to be accommodated automatically.
Association for Information and Image Management
In a collaborative effort, Mitek and SAIC have incorporated this new
technology into Mitek's industry-leading forms processing software.
This technology is being showcased at the annual conference and
exhibition of the Association for Information and Image Management
(AIIM).
Each year's AIIM Conference is the definitive event for the automated
document processing industry. This year's conference is being held in
Atlanta from the 11th through the 15th of April.
Thornton added: "Mitek and SAIC believe that the new capabilities
provided by this unique technology represent a fundamental advance in
automated document processing. This advance is anticipated to enable a
wide range of new applications. We are using the AIIM conference as a
venue for exploring the market potential of this new advance."
SAIC is the largest employee-owned high-technology company in the
United States. With more than 35,000 employees and estimated annual
revenues in excess of $4 billion, SAIC is ranked No. 347 in the Fortune
500.
Mitek Systems is a worldwide leader in intelligent character
recognition technology, products and services for the document imaging
markets, with headquarters at 10070 Carroll Canyon Rd., San Diego,
Calif. 92131.
The company's document imaging software engine and intelligent
character recognition products and services are used in
mission-critical and high-volume data entry, data capture and forms
processing applications installed in many of the world's leading
financial institutions, manufacturing and service companies, and
government organizations.
For more information about Mitek Systems, contact the company at
619/635-5900 or visit the company's Web site at www.miteksys.com.
Forward-Looking Statement Disclosure
With the exception of historical matters, the matters discussed in this
news release are forward-looking statements that involve risks and
uncertainty. Forward-looking statements include, but are not limited
to, statements relating to the development and pace of sales of the
company's products, expected trends and growth in the company's results
of operations, projections concerning the company's available cash flow
and liquidity, anticipated penetration in new and existing markets for
the company's products and the size of such markets, anticipated
acceptance of the company's products by existing and new customers, the
ability of the company to sustain any growth in sales and revenue and
the increase in sales representatives and other personnel. The
company's actual results could differ from such forward-looking
statements. There can be no assurance that the company will achieve the
results set forth herein.
Note to Editors: Mitek is a registered trademark of Mitek Systems.
-0- WAM/la* LS/la
CONTACT: Mitek Systems Inc., San Diego
John Thornton/William Boersing, 619/635-5900
or
Financial Relations Board, Los Angeles
310/442-0599
Stacy Roughan (General Info)
Moira Conlon (Investor Contact)
Mary Tokita (Media Contact)
KEYWORD: CALIFORNIA GEORGIA
INDUSTRY KEYWORD: COMPUTERS/ELECTRONIC TELECOMMUNICATIONS COMED
PRODUCT TRADESHOW Today's News On The Net - Business Wire's full file
on the Internet
with Hyperlinks to your home page.
URL: businesswire.com
*** end of story ***
|
