Is this page helpful?
[SerializableAttribute()]
public struct OcrCharacter
To get the recognized characters of a page, call IOcrPage.GetRecognizedCharacters after IOcrPage.Recognize.
To update the recognized characters of a page, call IOcrPage.SetRecognizedCharacters before calling IOcrDocument.Save, IOcrDocument.SaveXml.
IOcrPageCharacters implements the standard IList<T>, ICollection<T> and IEnumerable<T> interfaces with items of type IOcrZoneCharacters. Each item in the IOcrPageCharacters contains a collection of the character collections of the zones.
The IOcrZoneCharacters interface contains a collection of the characters for a particular zone.
IOcrZoneCharacters also implements IList<T>, ICollection<T> and IEnumerable<T> interfaces but with items of type OcrCharacter. Each item in the IOcrZoneCharacters contains a collection of the characters of the zone.
The IOcrPageCharacters interface also contains the IOcrPageCharacters.UpdateWord method that allow to modify the OCR recognition results by updating or deleting the words before optionally saving the results to the final output document.
For example, if you are interested in iterating through the characters of the 2nd zone in the page, you can do the following:
// Get the page characters
IOcrPageCharacters pageCharacters = ocrPage.GetRecognizedCharacters();
// Get the 2nd zone characters. Note, index is zero-based so 2nd zone is index 1
// You can also iterate through the pageCharacters collection and fine the IOcrZoneCharacters item with ZoneIndex = 1
IOcrZoneCharacters zoneCharacters = pageCharacters.FindZoneCharacters(1);
// Loop through the characters
foreach(OcrCharacter ocrCharacter in zoneCharacters)
{
// Do something with ocrCharacter
}
OcrCharacter is the most detailed information available about the recognized characters.
Touching characters, those whose shapes are physically joined in the page passed to the OCR engine, will result in a separate OcrCharacter structure for each recognized character within the block. However, the coordinate property of these characters (Bounds) will have identical coordinates defining a rectangle boundary for the character block. The order of the OcrCharacter structures representing a character block gives the order of the touching characters on the original document. This means the coordinates do not give information on the order of characters inside a boundary block.
To create an OcrCharacter object with all values initialized to default values, use OcrTypeManager.CreateDefaultOcrCharacter. This is a shortcut for languages that do not automatically initialize all the members of a structure, such as JavaScript.
using Leadtools;
using Leadtools.Codecs;
using Leadtools.Ocr;
using Leadtools.Forms.Common;
using Leadtools.Document.Writer;
using Leadtools.WinForms;
using Leadtools.Drawing;
using Leadtools.ImageProcessing;
using Leadtools.ImageProcessing.Color;
public void RecognizedCharactersExample()
{
// Create an image with some text in it
RasterImage image = new RasterImage(RasterMemoryFlags.Conventional, 640, 200, 24, RasterByteOrder.Bgr, RasterViewPerspective.TopLeft, null, IntPtr.Zero, 0);
Rectangle imageRect = new Rectangle(0, 0, image.ImageWidth, image.ImageHeight);
IntPtr hdc = RasterImagePainter.CreateLeadDC(image);
using (Graphics g = Graphics.FromHdc(hdc))
{
g.SmoothingMode = System.Drawing.Drawing2D.SmoothingMode.HighQuality;
g.FillRectangle(Brushes.White, imageRect);
using (Font f = new Font("Arial", 20, FontStyle.Regular))
g.DrawString("Normal line", f, Brushes.Black, 0, 0);
using (Font f = new Font("Arial", 20, FontStyle.Bold))
g.DrawString("Bold, italic and underline", f, Brushes.Black, 0, 40);
using (Font f = new Font("Courier New", 20, FontStyle.Regular))
g.DrawString("Monospaced line", f, Brushes.Black, 0, 80);
}
RasterImagePainter.DeleteLeadDC(hdc);
string textFileName = Path.Combine(LEAD_VARS.ImagesDir, "MyImageWithTest.txt");
string pdfFileName = Path.Combine(LEAD_VARS.ImagesDir, "MyImageWithTest.pdf");
// Create an instance of the engine
using (IOcrEngine ocrEngine = OcrEngineManager.CreateEngine(OcrEngineType.LEAD))
{
// Start the engine using default parameters
ocrEngine.Startup(null, null, null, LEAD_VARS.OcrLEADRuntimeDir);
// Create an OCR page
IOcrPage ocrPage = ocrEngine.CreatePage(image, OcrImageSharingMode.AutoDispose);
// Recognize this page
ocrPage.Recognize(null);
// Dump the characters into a text file
using (StreamWriter writer = File.CreateText(textFileName))
{
IOcrPageCharacters ocrPageCharacters = ocrPage.GetRecognizedCharacters();
foreach (IOcrZoneCharacters ocrZoneCharacters in ocrPageCharacters)
{
// Show the words found in this zone. Get the word boundaries in inches
ICollection<OcrWord> words = ocrZoneCharacters.GetWords();
Console.WriteLine("Words:");
foreach (OcrWord word in words)
Console.WriteLine("Word: {0}, at {1}, characters index from {2} to {3}", word.Value, word.Bounds, word.FirstCharacterIndex, word.LastCharacterIndex);
bool nextCharacterIsNewWord = true;
for (int i = 0; i < ocrZoneCharacters.Count; i++)
{
OcrCharacter ocrCharacter = ocrZoneCharacters[i];
// Capitalize the first letter if this is a new word
if (nextCharacterIsNewWord)
ocrCharacter.Code = Char.ToUpper(ocrCharacter.Code);
writer.WriteLine("Code: {0}, Confidence: {1}, WordIsCertain: {2}, Bounds: {3}, Position: {4}, FontSize: {5}, FontStyle: {6}",
ocrCharacter.Code,
ocrCharacter.Confidence,
ocrCharacter.WordIsCertain,
ocrCharacter.Bounds,
ocrCharacter.Position,
ocrCharacter.FontSize,
ocrCharacter.FontStyle);
// If the character is bold, make it underline
if ((ocrCharacter.FontStyle & OcrCharacterFontStyle.Bold) == OcrCharacterFontStyle.Bold)
{
ocrCharacter.FontStyle |= OcrCharacterFontStyle.Italic;
ocrCharacter.FontStyle |= OcrCharacterFontStyle.Underline;
}
// Check if next character is the start of a new word
if ((ocrCharacter.Position & OcrCharacterPosition.EndOfWord) == OcrCharacterPosition.EndOfWord ||
(ocrCharacter.Position & OcrCharacterPosition.EndOfLine) == OcrCharacterPosition.EndOfLine)
nextCharacterIsNewWord = true;
else
nextCharacterIsNewWord = false;
ocrZoneCharacters[i] = ocrCharacter;
}
}
// Replace the characters with the modified one before we save
ocrPage.SetRecognizedCharacters(ocrPageCharacters);
}
// Create an OCR document so we can save the results
using (IOcrDocument ocrDocument = ocrEngine.DocumentManager.CreateDocument(null, OcrCreateDocumentOptions.AutoDeleteFile))
{
// Add the page and dispose it
ocrDocument.Pages.Add(ocrPage);
ocrPage.Dispose();
// Show the recognition results
// Set the PDF options to save as PDF/A text only
PdfDocumentOptions pdfOptions = ocrEngine.DocumentWriterInstance.GetOptions(DocumentFormat.Pdf) as PdfDocumentOptions;
pdfOptions.DocumentType = PdfDocumentType.PdfA;
pdfOptions.ImageOverText = false;
ocrEngine.DocumentWriterInstance.SetOptions(DocumentFormat.Pdf, pdfOptions);
ocrDocument.Save(pdfFileName, DocumentFormat.Pdf, null);
// Open and check the result file, it should contain the following text
// "Normal Line"
// "Bold And Italic Line"
// "Monospaced Line"
// With the second line bold and underlined now
}
// Shutdown the engine
// Note: calling Dispose will also automatically shutdown the engine if it has been started
ocrEngine.Shutdown();
}
}
static class LEAD_VARS
{
public const string ImagesDir = @"C:\LEADTOOLS22\Resources\Images";
public const string OcrLEADRuntimeDir = @"C:\LEADTOOLS22\Bin\Common\OcrLEADRuntime";
}
IOcrPage.SetRecognizedCharacters
IOcrPage.GetRecognizedCharacters
OcrCharacter Structure