Convert an Image to Grayscale PDF for Better OCR in C#

Apryse

image

pdf conversion

ocr

pdf

Pre-processing is a critical step in improving the accuracy and quality of OCR (and also for ICR of handwriting) results.

One aspect of pre-processing involves de-skewing, and removal of artifacts such as staples and punched holes. Another is adjusting the color used by the document. OCR may give different results for text data extraction depending on whether a document is black and white, grayscale or color.

The Apryse SDK makes it easy to convert images to grayscale then import them into a PDF. In this article we will look at how to do this using C#, but the SDK also supports many other programming languages, and can be used on Windows, Linux and macOS.

General Process for Converting an Image to a Grayscale PDF

Copied to clipboard

Create PDF from source image.
Render page using PDFDraw.
Force grayscale color space.
Export JPEG.
Re-import image.
Save resulting PDF.

PDFNet needs to be initialized with a license key before it is used, and should be terminated when you have finished with it.

C#

Copied to clipboard

using pdftron; 
using pdftron.Common; 
using pdftron.PDF; 
using pdftron.SDF; 
 
namespace ConsoleApp1 
{ 
    internal class Program 
    { 
        static void ProcessFile(string[] args) 
        { 
            try { 

                string inputPath = "path_to_input_image"; 
                string exportGrayScalePath = "grayscale.jpg"; 
 
                PDFDoc doc = new PDFDoc(); 
                Page page = WritePage(doc, inputPath); 
                PDFDraw draw = new PDFDraw(300); 
                ObjSet hint_set = new ObjSet(); 
                Obj gray_hint = hint_set.CreateDict(); 
                gray_hint.PutName("ColorSpace", "Gray"); 
                
              //export as JPG, other options exist
                draw.Export(page, exportGrayScalePath, "JPEG", gray_hint); 
 
                doc.Close(); 
 
                PDFDoc grayDoc = new PDFDoc(); 
                WritePage(grayDoc, exportGrayScalePath); 
 
                // Save the PDF 
                grayDoc.Save("grayscale.pdf", SDFDoc.SaveOptions.e_linearized); 
                grayDoc.Close(); 
 
                Console.WriteLine("Image converted to grayscale and added to PDF successfully!"); 
 
            } 
            catch (PDFNetException e) 
            { 
                Console.WriteLine(e.Message); 
            } 
        } 
 
        static Page WritePage(PDFDoc doc, string imageFile)  
        { 
            ElementBuilder builder = new ElementBuilder(); 
            ElementWriter writer = new ElementWriter(); 
            Image img = Image.Create(doc, imageFile); 
            Page page = doc.PageCreate(new Rect(0, 0, img.GetImageWidth(), img.GetImageHeight())); 
            writer.Begin(page); 
            Element element = builder.CreateImage(img, 0, 0, img.GetImageWidth(), img.GetImageHeight()); 
            writer.WritePlacedElement(element); 
            writer.End(); 
            doc.PagePushBack(page); 
 
            return page; 
        } 
    } 

  [System.STAThread]
  static void Main(string[] args)
  {
    PDFNet.Initialize(APRYSE_LICENSE_KEY);
    ProcessFile();
    PDFNet.Terminate();
  }
}

Keep your license key confidential
License keys are uniquely generated and strictly confidential. Don't publish or store them in any public location, including public GitHub repositories.
You can get a trial license key for testing purposes, but you will need a commercial license in a production environment.

What next?

Copied to clipboard

You can now take the grayscale PDF and feed that into an OCR engine, or use it for any other purpose that suits your business needs.

Next Steps

Copied to clipboard

If you’re interested in trying Apryse SDK, you can start your trial right away. If you have any questions, please contact sales or check out our Discord.

Learn how to convert JPEG to PDF in C#.