For over 25 years, a name you can trust in the Data collection Industry!

SCANCORPORATION.com

Automated Data Solution Specialists

Food Service, Healthcare, Manufacturing, P.O.S. Retail, 
Utilities & Banking, Inventory Control
 
"It's all in the...Scan!"

Custom OCR Software

 

Customized Software Design is the process of converting your business needs or ideas into a computer software package to help your hardware product become fully functional or more efficient. We accomplish this by using a wide range of software development tools and highly-trained Development Engineers. We try to analyze your needs by working hand-in-hand with our clients to develop a software package that transforms your hardware specifications into a simple, easy, effective windows-based interactive program.  We specialize in designing OCR/Barcode programs that will...

 

  • help reduce manual user data entry, saving time and money
  • work with specialized and unique OCR hardware devices
  • integrate with various types of third-party software applications
  • design easy-to-use custom screens and pages
  • incorporate user-friendly help screens and manuals

 

 

 

SCAN's decoding library is a general-purpose library that processes Optical Character Recognition and barcode decoding. It can be ported to any embedded process with a C++ environment. Given a grayscale input image, the library will cut out and decode zones and iterate through multiple document definition, returning the results in strings. With a wide variety of image processing functions, the library may be tuned for maximum results. The library can be shut down to return all dynamically allocated memory to the system. Typically, the library is started up with an API call that allocates internal structures. It is then configured by importing a configuration string or by making individual API calls. After configuration, any number of decoding calls may be made without additional configuration.

 



SCAN Embedded Recognition Datasheet


 

♦ Features

 

 

OCR Symbologies 

  • OCR-A, OCR-B, E13-B, Handprint Arial, Courier, Times, U.S. currency

 

Barcode 

  • PDF 417, Code 39, Code 128, Codabar, UPC, EAN, JAN, Interleaved 2 of 5

 

Document Structure

  • Multiple Document Definitions

  • Auto-discrimination between documents

  • Multiple decoding zones within a document

 

Validator

  • Document verification against �rules� strings

  • Output formatter

 

Configuration Generation

  • Windows configuration Program

  • Graphic intermediate result debugging feedback

  • Regression Test on sample image database

  • Image Archiving

  • API code snippet generator

  • Click and drag zone generation

  • Configuration string generation

  • Input image formats � jpeg, bmp, limited PCX and TIFF

 

Image Processing

  • Adjustable threshold

  • Low pass filtering

  • Black space removal

  • Auto-document centering

  • Image or zone rotation and flipping through 90 degree angles

  • Small speckle removal

  • Large blob removal

  • Blob touching zone edge removal

  • Break joined characters

  • Voting on multiple passport recognition algorithms

 

Programming Interface

  • ASCII programming strings or individual API function calls

 

API interface

  • 'C' callable functions

 

Input Image

  • Grayscale - one byte per pixel

 

Output

  • String results for each zone
  • Strings for travel document fields
  • Error code
  • Error string

 

 

♦ Applications

 

 

  • Check Readers
  • Passport Readers
  • Remittance Processing 

 

 

♦ Specifications

 

 

System requirements

  • Code footprint 1 megabyte
  • Minimum RAM required 4 megabytes
  • C++ run-time environment

 

Development Requirements

  • C++ compiler
  • Method of getting image into and text out of the embedded system for regression testing

 

Accuracy

  • OCR-A, OCR-B, E13-B, 99.99% on ANSI Y-Grade media

 

Speed

  • 500 milliseconds (two line passport 200 MIPS Processor)