Aspose.OCR fails to read simple JPEG files

Question

I am testing Aspose.OCR, attempting to OCR a simple document, but finding that the OcrEngine.Process() returns jibberish with both my sample document and the sample provided by Aspose.

My code:

var license = new License();
license.SetLicense("Aspose.OCR.lic");
OcrEngine ocrEngine = new OcrEngine();
string text = null;
ocrEngine.Image = ImageStream.FromFile("Sample.Aspose.jpg");
if (ocrEngine.Process())
{
    text = ocrEngine.Text.ToString();
}
Assert.IsTrue(text.Contains("TRUTH"), text);

The Sample.Aspose.jpg is a copy of Aspose's GIT sample.

The text returned (truncated for brevity) starts with:

Avi [hhhBuyahLITITI Ll r h u -- - ] ---hhh --III-f LIII-fhh l t} ITI r F8 4 1 T Y L h IiRlm'kpfan order 081Dec - - hh - hh - - h - h j : t ITI lblel tljehrerlly }}ollnatffst/t trun IT IT } li IIIckaigf nigh ''I.. } : :;;.et}: fc.'IL:ef:t;;e;atc{1';;;:L IT':c:, ,.,.:,, ., ,...,. ''I

Equivalent jibberish is returned from a sample GIF.

Am I missing some simple settings for the OcrEngine?

Answer 1

The sample file that you have used is an example for OMR operation. You may consider the file “Sample.bmp” for OCR example. The code snippet is fine. It will work.

I work with Aspose as Developer evangelist.