Ocr not all columns of newspaper page

Aspose.OCR Product Family
1

Hello,

i want ocr from an image of scanned newspaper page, but the result is not satisfying. Is there an option for a better result? I have tested alle options of DetectAreasMode.

Here is the .net code

// Initialize the licensing object
Aspose.OCR.License license = new Aspose.OCR.License();
// Load the license from file
license.SetLicense(“Aspose.OCRfor.NET.lic”);
Aspose.OCR.AsposeOcr recognitionEngine = new Aspose.OCR.AsposeOcr();
// Recognize German text
Aspose.OCR.RecognitionSettings recognitionSettings = new Aspose.OCR.RecognitionSettings();
recognitionSettings.Language = Aspose.OCR.Language.Deu;
// Add images to OcrInput object
Aspose.OCR.OcrInput input = new Aspose.OCR.OcrInput(Aspose.OCR.InputType.SingleImage);
input.Add(“test.jpg”);
// Recognize image
List<Aspose.OCR.RecognitionResult> results = recognitionEngine.Recognize(input, new RecognitionSettings
{
DetectAreasMode = DetectAreasMode.DOCUMENT
});
// Save results
Aspose.OCR.AsposeOcr.SaveMultipageDocument(“test.pdf”, Aspose.OCR.SaveFormat.Pdf, results);
Aspose.OCR.AsposeOcr.SaveMultipageDocument(“test.txt”, Aspose.OCR.SaveFormat.Text, results);

Regards
Karsten

2

little changes of code
Aspose.OCR.RecognitionSettings recognitionSettings = new Aspose.OCR.RecognitionSettings();
recognitionSettings.Language = Aspose.OCR.Language.Deu;
recognitionSettings.DetectAreasMode = DetectAreasMode.DOCUMENT;

List<Aspose.OCR.RecognitionResult> results = recognitionEngine.Recognize(input, recognitionSettings);

3

@mozclaas

Would you please share your sample image for our reference as well? We will log an investigation ticket and share the ID with you.