C# OCR library for .NET applications

When people typically think of OCR (Optical Character Recognition), the first association is often with a scanner as the primary capture device. This association has historical reasons and is still prevalent in many contexts, providing consistent and controlled environment for capturing printed text from physical documents with unmatched quality. However, a scanner is specialized equipment that is not always at hand and requires a stationary workstation to operate. Fortunately, the modern world provides a convenient alternative to traditional scanners - a smartphone camera. The advancements in smartphone camera technology ensure that even an entry-level smartphone provides sufficient quality to capture OCR-ready documents. And built-in memory makes it easier than ever to digitize large quantities of documents, newspapers, books, street signs and other text on the go. All you need is the right technology to convert those photos into machine-readable text.

Our C# OCR library is specifically designed to recognize all types of images out-of-the-box and can be further fine-tuned to handle even the low-quality photos. Combined with a modern smartphone, it allows you to create powerful OCR applications for most everyday scanning and text recognition tasks. The most advanced image processing and document structure analysis are performed in a few lines of code, allowing you to focus on business rather than complex mathematical algorithms, neural networks and other technical intricacies.

// Configure preprocessing filters
PreprocessingFilter filters = new PreprocessingFilter {
  PreprocessingFilter.ContrastCorrectionFilter(),
  PreprocessingFilter.AutoDewarping()
};

// Add a photo for recognition
OcrInput photos = new OcrInput(InputType.SingleImage, filters);
photos.Add("photo.png");

// Fine-tune recognition setings
RecognitionSettings settings = new RecognitionSettings();
settings.Language = Language.Eng;
settings.DetectAreasMode = DetectAreasMode.CURVED_TEXT;

// Extract text from a page
AsposeOcr api = new AsposeOcr();
List<RecognitionResult> results = api.Recognize(photos, settings);

// Automatically correct spelling (English)
string text = results[0].GetSpellCheckCorrectedText(SpellCheckLanguage.Eng);
// Display recognized text
Console.WriteLine(text);

PDF is one of the most popular formats for scanning paper documents, especially due to its ability to combine multiple pages into a single file. This format is widely used for the exchange of contracts, invoices, legal documents, passports and ID cards, and many other documents between individuals, businesses, banks and government agencies. However, any scanned PDF is essentially a collection of images. It does not contain machine-readable text, so users cannot search, copy, or otherwise manipulate the document content.

Aspose .NET OCR offers you a fast, easy and highly reliable way to convert any scanned PDF into a fully searchable and indexable document. It accurately recognizes page content, converting it into a machine-readable text layer over the original image that can be selected, copied, read by text-to-speech software, and even automatically processed by translators, summarizers, and other AI-powered analytics tools.

// Load the scanned PDF
OcrInput pdf = new OcrInput(InputType.PDF);
pdf.Add("Delivery-Agreement.pdf");

// Recognize the text from document
AsposeOcr api = new AsposeOcr();
List<RecognitionResult> result = api.Recognize(pdf);

// Save searchable PDF
AsposeOcr.SaveMultipageDocument("Readable-Contract.pdf", SaveFormat.Pdf, result);
// Report progress
Console.WriteLine($@"Recognition finished. See '{Directory.GetCurrentDirectory()}\Readable-Contract.pdf'.");

Digital archives, especially in large organizations, often consist of a vast collection of scans and photos, many of which may contain multi-page documents. Efficient management and organization of such archives effectively is essential for easy information retrieval and navigation. However, images do not contain machine-readable text, making it impossible to search and analyze document content.

The C# OCR library allows you to easily search for text in images, regardless of the font, text size, style, and other parameters. The library also supports case-insensitive searches and regular expressions, which be extremely useful in various applications and industries. This functionality can be used for categorizing documents based on the content, keywords, or patterns found in the text; searching for specific terms or clauses within agreements and contracts; reorganizing files based on keywords or content found within them; locate and identify personal data within documents, making it easier to ensure GDPR compliance and manage sensitive information more effectively. Searching withing images also allows for creating automated workflows and streamline various business processes upon receiving signed contracts and invoices.

string sourceFolder = "images";
string searchFor = "OCR";

// Search for text in images
AsposeOcr api = new AsposeOcr();
foreach(var image in Directory.GetFiles(sourceFolder,"*.png"))
{
  bool found = api.ImageHasText(image, searchFor);
  if(found) Console.WriteLine($@"Found ""{searchFor}"" in image ""{image}""");
}