VintaSoft OCR .NET Plug-in is the add-on for VintaSoft Imaging .NET SDK, which allows to recognize text from images and save the recognition results to a text file or searchable PDF document.

Plug-in features

• The plug-in can be used for developing .NET 9, .NET 8, .NET 7 and .NET 6 applications (console, WPF, WinForms, ASP.NET Core, service, etc) for Windows
• The plug-in can be used for developing .NET 9, .NET 8, .NET 7 and .NET 6 applications (console, ASP.NET Core, service, etc) for Linux
• The plug-in can be used for developing .NET Framework 4.8, 4.7, 4.6, 4.5, 4.0 and 3.5 applications (console, WPF, WinForms, ASP.NET MVC, ASP.NET WebForms, service, etc) for Windows
• The plug-in core is written in C#
• AnyCPU, x86 and x64 mode support
  
• Cleanup document image before text recognition - VintaSoft Document Cleanup .NET Plug-in is necessary
  • Detect inverted text in document image and invert it
  • Remove halftones from document image
  • Clear border in document image
  • Remove hole punches in document image
  • Despeckle document image
  • Deskew document image
  • Detect text orientation
  
  
• Segment document image before text recognition
  • Detect segments on document image - VintaSoft Document Cleanup .NET Plug-in is necessary
  • Specify segments on document image from the code
  • Use mouse for selecting segments on document image in Image Viewer - this can be done using the RecognitionRegionEditorTool class from OCR Demo
  
  
• Specify the text recognition parameters
  • Specify the language for the whole document or for each recognition region
  • Supported more than 60 languages: English, German, French, Spanish, Portuguese, Russian, Italian, Dutch, Chinese, Arabic, etc. The full list of supported languages can be found here.
  • Specify the custom parameters of Tesseract OCR
  • Specify the "white list" of characters
  
  
• Recognize typed text from image or image collection using Tesseract OCR
  • Recognize text from image collection, the whole image or image region
  • Full Unicode support
  • Indicate progress of text recognition
  • Cancel text recognition if necessary
  • Obtain the recognition result in a hierarchy: document, page, region, paragraph, line, word, character
  • Report confidence, location, orientation, text and writing direction of each recognized paragraph, line, word and character
  
  
• Recognize a line of handwritten digits from image using neural network
  • Recognize a line of handwritten digits from image using neural network
  • Report confidence, location, value of each recognized digit
  
  
• Edit text recognition results
  • Edit text recognition results from the code
  • Use mouse for editing text recognition results in Image Viewer - this can be done using the OcrResultEditorTool class from OCR Demo
  • Import text recognition result from hOCR format
  
  
• Save text recognition results
  • Save recognition results as text or formatted text
  • Save recognition results to a searchable PDF document in 3 modes (text over image, image over text, text only) - VintaSoft PDF .NET Plug-in is necessary
  • Export text recognition result into hOCR format

Development and Deployment requirements

• Development requirements:

  • Development environments: Microsoft Visual Studio, Microsoft Studio Code, JetBrains Rider, .NET CLI
  • Programming languages: C#, VB.NET, any .NET compatible language
  • Development platforms: .NET, WinForms, WPF, ASP.NET
  
  

• Deployment requirements:

  • Windows (Windows 11/10/8.1/8/7; Windows Server 2025/2022/2019/2016/2012/2008; 32-bit and 64-bit) with .NET 9, .NET 8, .NET 7 or .NET 6 and Microsoft Visual C++ 2019 Redistributable Package
  • Linux (Ubuntu 24/23/22/20/18; Debian 12/11/10; Fedora 40/39/38/37/36/35; OpenSUSE 15; CentOS 9/8/7; Redhate Enterprise Linux 9/8/7; Alpine 3) with .NET 9, .NET 8, .NET 7 or .NET 6
  • macOS (macOS 14/13/12/11/10) with .NET 9, .NET 8, .NET 7 or .NET 6