OCR & Document Conversion Processing Profiles

What are predefined Processing Profiles and what is their role in the development process

Predefined Processing Profiles are code samples, that can be used to speed own development. The processing profiles contain processing options that are optimized for individual processing scenarios - offering different pre-set options for each scenario.

ABBYY introduced new Processing Profiles with the version 10 of FineReader Engine to help software developers to speed up the development process. Software engineers that are less experienced in the area of OCR can as well use these ready-made profiles instead of own testing and finetuning the application and its OCR processing parameters.

Predefined Processing Profiles for faster development

Processing profiles can be used to speed up own development. The ABBYY FineReader Engine is a comprehensive SDK with:

  • more than 3.000 API calls, objects and parameters
  • extensive help file containing over 700 pages
  • 100+ essential parameters for opening, pre-processing, analysis, recognition and exporting of a document

Using the predefined Processing Profiles allows to implement optimized parameters for specific (most common) processing scenarios.

Following Processing Profiles are available:

  • Document/book conversion for archiving (in 2 modes: speed & accuracy)
  • Document conversion for content reuse (in 2 modes: speed & accuracy)
  • Ground text extraction for fields detection and document classification (in 2 modes: speed & accuracy)
  • Field level recognition
  • Barcode recognition (in 2 modes: speed & accuracy) (New speed mode introduced in FineReader Engine 11 R4, Nov 2014)
  • High Compressed Image Only Pdf — for creating high-compressed PDF files which contain entire documents saved as pictures.
  • Business Cards Processing — for recognizing business cards
  • Engineering Drawings Processing — for recognizing technical drawings, with text arranged in different directions
  • Backward compatibility

Simple to use:

[STAThread]
static void Main()
{
          // Loading ABBYY FineReader Engine (GetEngineObject)
          engine.LoadPredefinedProfile("DocumentConversion_Accuracy");
          // Possible profile names are:
          //   "DocumentConversion_Accuracy", "DocumentConversion_Speed",
          //   "DocumentArchiving_Accuracy", "DocumentArchiving_Speed",
          //   "BookArchiving_Accuracy", "BookArchiving_Speed",
          //   "TextExtraction_Accuracy", "TextExtraction_Speed",
          //   "FieldLevelRecognition",
          //   "BarcodeRecognition_Accuracy", "BarcodeRecognition_Speed",
          //   "HighCompressedImageOnlyPdf",
          //   "BusinessCardsProcessing",
          //   "EngineeringDrawingsProcessing",
          //   "Version9Compatibility"
          // Loading and preprocessing of images
          // Document recognition (IFRDocument::Process)
          // Document export (IFRDocument::Export)
          // Unloading ABBYY FineReader Engine (DeinitializeEngine)
}

Learn more in this video (valid for FineReader Engine 11 as well as higher versions):

 

Continous tuning of Processing Profiles by ABBYY software engineers

ABBYY is continuously improving its core recognition technologies and with any change, the processing profiles are automatically optimized and thoroughly tested by the ABBYY development team.
The ABBYY automated testing framework includes:
    • Measurement tools
    • Ground truth preparation tools
    • Testing image batches and marking out
    • Option to test different ABBYY OCR engines

abbyy_testing_framework_illu.png

 

Have more questions? Submit a request

Comments

0 comments

Please sign in to leave a comment.