OpenCV Character Recognition

A Windows Forms application for OCR text extraction from images using EmguCV and Tesseract.

About This Project

This project was originally created ten years ago and remained unchanged since then. In 2025 I decided to modernize it and remove the cumbersome manual EmguCV/OpenCV installation process. The new version uses NuGet packages instead, making setup much simpler.

Features

• OCR text extraction from images
• Support for multiple image formats (JPG, PNG, GIF, BMP)
• Modern async/await patterns for responsive UI
• No manual environment setup required
• All dependencies managed through NuGet

Requirements

• Windows 10 or later
• .NET 10 SDK
• Visual Studio 2022 or VS Code with C# Dev Kit

Setup

No manual environment configuration needed. The build process automatically downloads required Tesseract language data.

git clone https://github.com/brakmic/OpenCV.git
cd OpenCV
dotnet build

The first build will download tessdata files automatically.

Additional Languages

By default, only English language data is downloaded. To use other languages, download them manually using the provided script.

Example for German:

pwsh -ExecutionPolicy Bypass -File scripts/Download-Tessdata.ps1 -TessdataPath CharacterRecognition/tessdata -Language deu

Then update the Language property in CharacterRecognition/appsettings.json to match your chosen language code.

Available language codes: eng, deu, fra, spa, ita, por, and many more.

Running the Application

dotnet run --project CharacterRecognition/CharacterRecognition.csproj

Or open OpenCV.sln in Visual Studio and press F5.

Usage

1. Start the application
2. Load one of the sample images from the Assets folder (these are invoice documents)
3. Click on Analyze and wait for the OCR task to complete
4. View the extracted text in the result window

Technical Details

• Framework: .NET 10
• UI: Windows Forms
• OCR Engine: Tesseract 4 via EmguCV 4.12
• Architecture: x64

License

MIT