IDA Recognition

IDA Recognition

Unmatched OCR & ICR accuracy

Unmatched OCR & ICR accuracy

Recognition_Gif

Main Feature for Data Capture

Despite scenarios like distorted, poor-quality scans with machine-print and difficult-to-read handwriting, IDA Recognition delivers OCR and ICR results with exceptional accuracy. As a result, there is a reduced need for manual corrections, leading to enhanced straight-through processing.

We offer the Recognition Feature as an SDK, providing even greater flexibility and integration options.

Performance

Machine-print Accuracy

> 99 %

Handwriting Accuracy

> 95 %

Key Features

Patentiert
Patented PerceptionMatrix

IDA captures complex documents, preserving all transcriptions for high accuracy.

Wide scope of scenarios

IDA handles machine-printed and handwritten text, barcodes, and more.

Auto language detection

IDA recognizes multiple languages simultaneously, based on refined language models.

Comprehensive PDF capabilities

IDA stores SmartPDFs with text layers, supporting PDF/A for efficient archiving.

Versatile JSON output

Access captured documents easily for downstream tasks using exclusive JSON format.

Easy deployment and integration

On-premises or cloud-based as Java app or Docker container, integrating seamlessly with gRPC API.

Explore IDA Recognition

Available Languages

IDA Recognition offers character recognition for Western languages that use the Latin script. Pre-trained language models to further enhance the recognition results are provided for the following languages:

English

Spanish

French

Portuguese

German

Turkish

Italian

Dutch

Czech

Swedish

Norwegian

Slovene

  • Latin
  • Additional languages available by request

Recognition Add-ons

Customize IDA Recognition with a versatile range of Recognition Add-ons. These separately licensable extensions enhance the scope of IDA Recognition, encompassing functionalities such as capturing 1D & 2D codes, redaction, capuring of historical scripts, arabic and chinese character recognition, and advanced table extraction.

Barcode Recognition_Screenshot

Barcode Recognition

Supported codes:

1D

2D

  • CodaBar
  • Code39

  • Code93

  • Code128

  • EAN8
  • EAN13
  • ITF
  • UPC-E
  • Aztec
  • DataMatrix
  • PDF417
  • QR

Historical Scripts

  • Historical blackletter scripts (Gothic)
  • Historical handwriting (f.e. Kurrent, Sütterlin)

  • English, French, German, Latin, Italian
Historical Scripts_Screenshot
Entity Finder_Screenshot

Entity Finder

  • List of words

  • Regular expressions (RegEx)

  • Highlighting
  • Redaction

Non-Latin Script LanguagesBETA

  • Chinese (simplified)
  • Arabic
Non-Latin Script Languages_Image
Table Recognition_Image

Table RecognitionBETA

  • Interpretation of table structures

Ready for High-level OCR?

We would like to find our more about your business needs.