Documentation
Table of contents

Go to DCV Architecture

Semantic Processing

A semantic processing operation usually follows Text-line Recognition or Barcode Decoding:

flowchart LR;
     A[3.Text-line Recognition]-->C[Semantic Processing]
     B[3.Barcode Decoding]-->C[Semantic Processing]
     style C fill:#f96,stroke:#333,stroke-width:4px

What is Semantic Processing

The purpose of “Semantic Processing” is to make the recognized text more user friendly. At present, it means to parse a text, which follows a specific specification, into fields. For example, on this sample passport image

semantic-processing-sample image

we can find the MRZ and recognize the following text lines with a “Recognize-Text-Lines” task:

P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<
L898902C36UTO7408122F1204159ZE184226B<<<<<10

these lines don’t make much sense to us, so we parse them with a “Semantic Processing” task and with a bit more code, the result becomes something like this:

Document Type : PASSPORT (TD-3)
Issuing State : UTO
Surname : ERIKSSON
Given Name : ANNA MARIA
Passport Number : L898902C3
Nationality : UTO
Date of Birth (YYYY-MM-DD) : 1974-08-12
Gender : F
Date of Expiry (YYYY-MM-DD) : 2012-04-15

This page is compatible for:

Version 1.0

Is this page helpful?

YesYes NoNo

In this article:

latest version

    • Latest version (2.2.30)
    • Version 2.x
      • Version 2.2.20
      • Version 2.2.10
      • Version 2.2.0
      • Version 2.0.30
      • Version 2.0.20
      • Version 2.0.10
      • Version 2.0.0
    Change +
    © 2003–2024 Dynamsoft. All rights reserved.
    Privacy Statement / Site Map / Home / Purchase / Support