OCR: How to recognize CIJ characters from image in .NET

In This Topic

Because CIJ (Continuous Inkjet) characters are formed by tiny droplets, the Tesseract OCR engine struggles to find continuous edges.
If you want to recognize text, which contains CIJ characters, please process image before text recognition and "convert" CIJ characters into text characters with smooth lines.
CIJ characters can be "converted" into text characters with smooth lines using the morphological filters - first, dilate symbols on image (DilateCommand class); next, erode symbols on image (ErodeCommand class).

Here is C#/VB.NET code that shows how to "convert" text with CIJ characters into text characters with smooth lines and recognize text using Tesseract OCR engine:

/// <summary>
/// Recognizes CIJ characters from image using Tesseract OCR engine.
/// </summary>
/// <param name="filename">The name of file, which stores images with CIJ characters.</param>
public static void RecognizeCIJCharactersUsingTesseractOCR(string filename)
{
    // create an image collection
    using (Vintasoft.Imaging.ImageCollection images = 
        new Vintasoft.Imaging.ImageCollection())
    {
        // add images from file to the image collection
        images.Add(filename);

        System.Console.WriteLine("Create Tesseract OCR engine...");
        // create the Tesseract OCR engine
        using (Vintasoft.Imaging.Ocr.Tesseract.TesseractOcr tesseractOcr = 
            new Vintasoft.Imaging.Ocr.Tesseract.TesseractOcr())
        {
            System.Console.WriteLine("Initialize OCR engine...");
            // specify that English text should be recognized
            Vintasoft.Imaging.Ocr.OcrLanguage language = Vintasoft.Imaging.Ocr.OcrLanguage.English;
            // create the Tesseract OCR settings
            Vintasoft.Imaging.Ocr.Tesseract.TesseractOcrSettings settings =
                new Vintasoft.Imaging.Ocr.Tesseract.TesseractOcrSettings(language);
            // specify that image contains single text block
            settings.RecognitionRegionType = Vintasoft.Imaging.Ocr.RecognitionRegionType.RecognizeSingleBlock;
            // init the Tesseract OCR engine
            tesseractOcr.Init(settings);

            // for each image in image collection
            foreach (Vintasoft.Imaging.VintasoftImage image in images)
            {
                System.Console.WriteLine("Recognize the image...");

                // dilate image
                DilateCommand dilateCommand = new DilateCommand();
                dilateCommand.WindowSize = 5;
                dilateCommand.ExecuteInPlace(image);
                // erode image
                ErodeCommand erodeCommand = new ErodeCommand();
                erodeCommand.WindowSize = 3;
                erodeCommand.ExecuteInPlace(image);

                // recognize text in image
                Vintasoft.Imaging.Ocr.Results.OcrPage ocrResult = tesseractOcr.Recognize(image);

                // output the recognized text

                System.Console.WriteLine("Page Text:");
                System.Console.WriteLine(ocrResult.GetText());
                System.Console.WriteLine();
            }

            // shutdown the Tesseract OCR engine
            tesseractOcr.Shutdown();
        }

        // free images
        images.ClearAndDisposeItems();
    }
}

VB.NET

''' <summary>
''' Recognizes CIJ characters from image using Tesseract OCR engine.
''' </summary>
''' <param name="filename">The name of file, which stores images with CIJ characters.</param>
Public Shared Sub RecognizeCIJCharactersUsingTesseractOCR(filename As String)
    ' create an image collection
    Using images As New Vintasoft.Imaging.ImageCollection()
        ' add images from file to the image collection
        images.Add(filename)

        System.Console.WriteLine("Create Tesseract OCR engine...")
        ' create the Tesseract OCR engine
        Using tesseractOcr As New Vintasoft.Imaging.Ocr.Tesseract.TesseractOcr()
            System.Console.WriteLine("Initialize OCR engine...")
            ' specify that English text should be recognized
            Dim language As Vintasoft.Imaging.Ocr.OcrLanguage= Vintasoft.Imaging.Ocr.OcrLanguage.English
            ' create the Tesseract OCR settings
            Dim settings As Vintasoft.Imaging.Ocr.Tesseract.TesseractOcrSettings =
                New Vintasoft.Imaging.Ocr.Tesseract.TesseractOcrSettings(language)
            ' specify that image contains single text block
            settings.RecognitionRegionType = Vintasoft.Imaging.Ocr.RecognitionRegionType.RecognizeSingleBlock
            ' init the Tesseract OCR engine
            tesseractOcr.Init(settings)

            ' for each image in image collection
            For Each image As Vintasoft.Imaging.VintasoftImage In images
                System.Console.WriteLine("Recognize the image...")

                ' dilate image
                Dim dilateCommand As DilateCommand = New DilateCommand()
                dilateCommand.WindowSize = 5
                dilateCommand.ExecuteInPlace(image)
                ' erode image
                Dim erodeCommand As ErodeCommand = New ErodeCommand()
                erodeCommand.WindowSize = 3
                erodeCommand.ExecuteInPlace(image)

                ' recognize text in image
                Dim ocrResult As Vintasoft.Imaging.Ocr.Results.OcrPage = tesseractOcr.Recognize(image)

                ' output the recognized text

                System.Console.WriteLine("Page Text:")
                System.Console.WriteLine(ocrResult.GetText())
                System.Console.WriteLine()
            Next

            ' shutdown the Tesseract OCR engine
            tesseractOcr.Shutdown()
        End Using

        ' free images
        images.ClearAndDisposeItems()
    End Using
End Sub

Send Feedback