Zonal OCR or Field Level recognition is a type of Optical Character recognition that allows a user to scan and read specific zones of the image. Extracting that data faster and with a higher degree of accuracy is the goal of Zone OCR.

.NET OCR component can add accurate, fast & thread-safe OCR support to your .NET imaging application. The input image can be from you local image file, fax document, screenshots, or capture images from TWAIN.

This online guide shows you how to extract text from a specific area of an image using C# language.


// Please note:
// If you choose the x64 platform, please copy the "XsOCR_Tesseract.dll" and "XsOCR_Lept.dll"
// from the x64 folder to the same level path which "XsOCR.dll" in. 
// Otherwise, please copy from x86 folder.

// Create an OCR Engine instance
OCREngine engine = new OCREngine();
// Set the absolute path of tessdata
engine.DataPath = "F:/tessdata/";
// Set the target text language you want to recognize
engine.Language = "eng";
// Customized region can be set to OCR the text
engine.AreaToRecognize = new System.Drawing.Rectangle(50, 50, 150, 150);
// Recognize text from image file
string text = engine.DoOCR("F:/sample.jpg");

System.Console.WriteLine(text);

Find and Download all OCR language data from this page.

Notice - If you used the trial version of OCR SDK, the first character of result is symbol "?"