Über LEADTOOLS Forms Recognition Module

Programmieren von Hochgeschwindigkeits-Dokumentenscansoftware.

LEADTOOLS Forms Recognition is a high-level .NET SDK that harnesses the power of LEAD's image processing technology to intelligently identify document components and features that can be used to recognize and classify scanned documents. The LEADOOLS Forms Recognition toolkit can be used to process a list of images, return unique XML data that describes them, and then store and use that data set as a standard for recognizing future digital images. This approach is especially powerful when combined with LEADTOOLS scanning, deskew, document image clean up, OCR, ICR, OMR, MICR, redaction, and annotation functionalities for automating end-to-end (capture-to-data-processing) document handling workflows.

Key Features of LEADTOOLS Forms Recognition

Code Support

  • High level classes enable adding form recognition features with very little code
  • Includes full source code for an end-user Forms Recognition and Processing application, plus hundreds of lines of sample source code

Build Workflows

  • Use in conjunction with other LEADTOOLS Document Imaging functions to build workflows that incorporate scanning,  OCR, ICR (hand-written text), OMR (to extract check marks, check boxes, etc), MICR, redaction and annotations with recognition and processing
  • Automatic, manual (text-analytic), trained, and multi-level recognition
  • Flexible automatic recognition engine can be customized to use Barcode, OCR, line detection, and special object detection (such as logos, text blocks, and shapes)
  • Recognizes any of 150+ input file formats, including different bits-per-pixel, dpi, scales, skew angles, and noise

Forms, Region, and Data Support

  • Define multi-page forms for recognition and processing
  • Store any number of pre-defined master forms for recognition
  • Recognize forms at both form and page level, to compare an image to any form page or compare a complete multi-page form to any predefined master form
  • Load and save form identification information as XML for storage in a database
  • Recognize any number of regions or objects on a page
  • Define inclusion and exclusion regions to be used to classify forms
  • Supports a variety of data fields:
    • Text fields support both printed and hand-written text
    • Supports regular expressions to find / validate text in a specified format
    • Image fields to extract signatures, personal pictures, logos, and fingerprints
    • Barcode fields support all barcode 1D (Linear) and 2D types, such as PDF417, DataMatrix, QR, EAN, UPC, Databar, and 4-state

Results Enhancement

  • Auto-registration (deskew) and clean-up to improve recognition results
  • Manual feature weighting for improved recognition performance
  • Generates comprehensive results and confidence level reports to assess performance