For developers needing to extract text from images, scans and PDFs, IronOCR is a powerful C# OCR library supporting 127 languages. Built on a modern machine‑learning engine, IronOCR reads text from photographs, PDFs, screenshots and even specialized documents like passports or license plates. The engine returns structured results (plain text, hOCR, OCR data with coordinates) and can embed recognized text into searchable PDFs.

Key Features

• Multi‑language recognition: IronOCR supports dozens of languages, including complex scripts like Chinese and Arabic, as well as right‑to‑left languages. You can load multiple language packs simultaneously or train custom languages.
• Image preprocessing and filtering: The built‑in filter wizard lets you chain image filters (denoise, sharpen, contrast adjustment, histogram equalization) to improve recognition on low‑quality scans. Automatic orientation detection and deskew ensure text is aligned correctly.
• Barcodes and structured output: In addition to text, IronOCR detects QR and barcode symbologies. The API can return text with bounding boxes, confidence levels and layout information; this structured data is useful for forms processing and data extraction.
• Concurrency and performance: IronOCR supports multithreading and asynchronous calls, enabling high‑volume scanning on servers or in cloud functions. The library also caches fonts and dictionary data to reduce initialization time.
• Cross‑platform support: Signed binaries run on Windows, Linux and macOS, with support for .NET 10, 9, 8, 7, 6 and .NET Framework 4.6.2+. It works in web, desktop and mobile projects.
• NuGet & documentation: Install with Install‑Package IronOcr or visit NuGet to download. Documentation and tutorials are available at the IronOCR docs and examples page.

2025 Change‑Log Highlights

IronOCR’s October 2025 release updated internal dependencies and improved platform reliability. September 2025 optimized memory consumption for TIFF inputs and updated System.Drawing to version 2025.9.3. August 2025 introduced new orientation detection modes in DetectPageOrientation and improved recognition across rotated images. Earlier in May 2025, IronOCR added IronOcr.Extensions.AdvancedScan for Linux/macOS and introduced options to create searchable PDFs with or without filters (useful when generating small file sizes). These updates show the team’s focus on performance and cross‑platform parity.

Tutorials and Resources

Check out the Quickstart tutorial to learn how to load images, apply filters and extract text. Sample code on the examples page shows how to recognize a scanned document, read QR codes and generate a searchable PDF. For guidance on improving accuracy, the filter wizard documentation explains how to apply denoise, brightness, and inversion filters. A free 30‑day trial with technical support is available.