The C# OCR Library

Go beyond basic OCR. Get an advanced ML engine for .NET 10 and preprocessing that reads scans, screenshots, and handwriting.

Read images, PDFs, and structured documents

125+ languages and filters for poor quality scans

Export structured data, text, hOCR, and searchable PDFs

Free 30-day trial, full features, no credit card

Images to Text with OCR See All 38 Code Examples

using IronOcr;

var ocr = new IronTesseract();

using var input = new OcrInput();
input.LoadImage("attachment.png");
input.LoadPdf("report.pdf");

OcrResult result = ocr.Read(input);
string text = result.Text;

Install-Package IronOcr

Next Step: Start Free Trial

Fully functional without watermarks

30-day trial - no credit card required

Next Step: Start Free Trial

Fully functional without watermarks

30-day trial - no credit card required

PM > Install-Package IronOcr

IronOCR

See what’s new:

Added SaveAsSearchablePdf() Support

Cross Platform Support

Designed for C#, F#, & VB.NET running on .NET 10, 9, 8, 7, 6, 5, Core, Standard, or Framework

.NET Languages:

C#, VB.NET, F#

App Types:

Console, Web, & Desktop

Platforms:

.NET 10, 9, 8, 7, 6 & 5
.NET Standard 2
.NET Framework 4.6.2+

Operating Systems:

Windows
macOs
Linux
Docker
Azure
AWS

IDEs:

Microsoft Visual Studio
Jetbrains ReSharper & Rider

Binary Certification

Microsoft Authenticode

A C# OCR Library that prioritizes accuracy, ease of use, and speed.

IronOCR is the leading C# OCR library, offering superior text extraction from images and PDFs compared to Tesseract. Its user-friendly API enables developers to implement OCR in their .NET projects within minutes.

Get your free 30-day Trial Key instantly.

Your trial key should be in the email.
If it is not, please contact
support@ironsoftware.com

Tesseract OCR in the language you need. We support 125+.

View Full Features Set

Why IronOCR?

Accurate text extraction. Built for real-world documents. Works on Windows, Linux, macOS & cloud.

Industry-leading accuracy

Tesseract 5 & advanced ML engines → what humans can read, IronOCR can read.
Auto-detection: text regions, tables, barcodes. No manual configs needed.
Handles real documents with confidence: skewed scans, photos, low DPI.

Outcome: Extract usable text, not gibberish that requires manual cleanup.

Up & running in 5 mins

PM >
Install-Package IronOcr

➝ 5-minute quick starts.
Simple API: Read(), ReadPdf(), ReadBarcode(), and intuitive method names.
Smart defaults: Auto-preprocessing, language detection, layout analysis built-in.

Outcome: Ship OCR features this sprint, not next quarter.

Enterprise performance

.NET 10, Core, Standard, or Framework
Multi-threading: Process batches 5-10x faster than sequential.
Cloud/containers: Azure, AWS, Docker/Kubernetes, serverless.

Outcome: Scale from desktop tools to enterprise document processing.

Robust processing

125+ languages with proper RTL/vertical support.
Advanced ML to deskew, denoise, resolution enhancement.
Structured data: Tables to CSV/Excel, forms to JSON.

View Code Examples

Developer-first support & licensing

Perpetual license — buy once, use forever. Optional updates and renewals.
24/5 engineer support (24/7 in Premium). Real humans, median chat response <1 min.
30-day money-back guarantee. Not the right fit? Full refund, no questions.

Chat With an Engineer

For builders

Features you'll actually use:
- text/table extraction
- barcode/QR reading
- PDF to searchable PDF
- confidence scores
- handwriting OCR
- 125+ languages
- image preprocessing
- batch processing
- custom regions (ROI)
- document classification
Works everywhere: Windows, Linux, macOS, Docker, Azure, AWS, cloud

30 days, fully-functional product

For buyers

Trusted by 2M+ engineers • 14M+ NuGet installs.
Teams report 40% dev-time saved vs DIY/OSS wrappers.
Redistribution/SaaS options available.
Procurement-friendly: perpetual license, optional renewals, invoice available, multiple payment options (including local).
30-day money-back guarantee.

See Licensing

Trusted by millions of engineers worldwide

Every time developers ask, it's a YES! from IronOCR

switch(faqQuestions)
{ 
    case "Can I deploy my application using the library to Linux, macOS, or Windows, including Docker containers?": return "YES";
    case "Can I render pixel-perfect PDFs from complex HTML5, CSS3, and JavaScript?": return "YES";
    case "Is the library thread-safe for use in highly concurrent, multi-threaded server applications and APIs?": return "YES";
    case "Can I set custom viewport dimensions and print media CSS to control responsive layouts?": return "YES";
    case "Is IronPDF compatible with Azure Functions?": return "YES";
    case "Can I permanently redact sensitive text and images from PDF documents?": return "YES";
    case "Can I generate a PDF from a URL that requires authentication via custom headers, cookies, or form logins?": return "YES";
    case "Can I access PDF DOM through ObjectModel property?": return "YES";
    case "Can I programmatically fill, read, and flatten interactive AcroForm?": return "YES";
    case "Does it support generating PDF/A and PDF/UA compliant documents for archiving and accessibility standards?": return "YES";
    case "Can I secure a document with 256-bit AES password encryption and set user permissions?": return "YES";
    case "Can I apply and verify digital signatures on a PDF using X.509 certificates?": return "YES";
    case "Can I test IronPDF in my live project for free now?": return "YES, click here to start your 30-day free trial";
}