Name and Address Block Reader (NABR) [CedarTech, Inc]

The reading of names and addresses is one of the most complex tasks inautomated forms processing. The Name and Address Block Reader (NABR) isan integrated real-time system to read hand printed as well as machineprinted images containing name and address block ubiquitous in a varietyof documents. The NABR System is the most cost effective solution foryour document processing. In general, the scanned name and addressblock image is processed through the two major steps - Document Analysisand Document Recognition. During document analysis phase, applied are asuit of image processing techniques such as connected component analysis,skew detection and correction, address block extraction, label detectionand hand-print / machine-print discrimination. Once an image is analizedcompletely, all the necessary information is represented in a structuredin a structured fashion, so that the second phase of the process can beapplied. According to the hand-print / machine-print classificationresults, the image goes through a different streams. The key steps are:address parsing, character recognition and segmentation, word recognitionand name and address search and correction with postal database lookup(ZIP+4 and City-State-ZIP files) and name database. The system isdesigned with a SPARCStation(TM) [SPARC 10, SPARC 20 or UltraSPARC] withone monitor, one external disk drive and one tape drive. There are oneor multiple CPU's available according to the selection of the platform.The system can handle 2, to 8, name and address block images perhour. The I/O interface is 10Base-T or 100Base-T Ethernet through whichthe input image is fed and the output result is provided for futherprocess.

Source Avail: No
Product Special Handling: None.
Operating Systems: Solaris Sparc 2.5

CedarTech, Inc
5500 Main St
Williamsville, NY 14221
USA
Phone: 7165650420
Fax: 7165650308

Optical Character Recognition

Name and Address Block Reader (NABR)

CedarTech, Inc