Error: Recognition of image failed

[Hari.K]

Hi all,

I keep getting the error- "Recognition of image failed" for which I am unable to figure out the root cause.

What I am trying to do : I have a PDF document(12 pages) on which I am trying perform the OCR. 

Step-1: Splitting the entire PDF document to individual pdf pages.

Step-2: Converting each page to a Bitmap image. 

Step-3: Each Bitmap is now being fed as an input to the Tesseract so that it will perform the OCR and gives back the result.

Where I am getting the error: This error is not occuring consistantly in a specifc page. I am getting the error randomly at any pagenumber (i.e error can occur in any page say: 7 or 8 or 11 etc..)

Exact line of code where I'm getting the error: using (ResultIterator r = currentPage.GetIterator())

More Observations: I noticed that I am getting this error only when the PDF I uploaded is considerably big., like if the PDF is more that 10 pages or so..

Looking for some inputs from you guys.. 

Thanks in advance..

Code in high level:

            Tesseract.Page page;

    TesseractEngine ocr;

            

           public void SplitFile(string path)

        {

            Spire.Pdf.PdfDocument document = new Spire.Pdf.PdfDocument(path);

            System.Drawing.Bitmap bitmap;

            int pageNumber = 0;

            int pageCount = document.Pages.Count;

            try

            {

                for (int i = 0; i < pageCount; i++)

                {

                    bitmap = (Bitmap)document.SaveAsImage(pageNumber, PdfImageType.Bitmap, 450, 450); //450 is the DPI

                    BitmapToPixConverter b = new BitmapToPixConverter();

                    Tesseract.Pix pix = b.Convert(bitmap);                 

                    ProcessOCR(pix, documentId, pageNumber, pageCount);

                    ocr.Dispose();

                    pageNumber++;

                }

                document.Close();

            }

}

    public void ProcessOCR(Pix image, int pageNumber, int pageCount)

            {

                List<Coordinates> lstCoordinates;

                ocr = new TesseractEngine(HttpContext.Current.Server.MapPath(@"~/tessdata"),"eng",EngineMode.Default);

                using (page = ocr.Process(image, PageSegMode.SingleColumn))

                {

                    

                    lstCoordinates = GetWordsWithCoordinates(page, image.Width, image.Height, pageNumber);

                }

}

    public List<Coordinates> GetWordsWithCoordinates(Page currentPage, int width, int height, int pageNumber)

            {

                List<Coordinates> words = new List<Coordinates>();

            

                using (ResultIterator r = currentPage.GetIterator())

                {

                    do

                    {

                                           

                        string word = r.GetText(PageIteratorLevel.Word);

if (word != null)

                        {

// fetch the coordinates of the word and do something

}

   } while (r.Next(PageIteratorLevel.Word));

   }

}