Save Searchable PDFs in C# with IronOCR

Working with Multi-Page Documents

When dealing with PDF OCR operations on multi-page documents, IronOCR processes each page sequentially and maintains the original document structure. Here's an example of converting a multi-page scanned PDF into a searchable PDF:

using IronOcr;

// Initialize IronTesseract with configuration
var ocrTesseract = new IronTesseract
{
    Configuration = 
    {
        RenderSearchablePdf = true,
        PageSegmentationMode = TesseractPageSegmentationMode.Auto
    }
};

// Load a multi-page PDF
using var pdfInput = new OcrPdfInput("multi-page-scan.pdf");

// Optionally specify page range (e.g., pages 1-10)
pdfInput.SelectPages(1, 10);

// Perform OCR with progress tracking
OcrResult result = ocrTesseract.Read(pdfInput);

// Save as searchable PDF
result.SaveAsSearchablePdf("searchable-multi-page.pdf");

// Display total pages processed
Console.WriteLine($"Processed {result.Pages.Length} pages");

using IronOcr;

// Initialize IronTesseract with configuration
var ocrTesseract = new IronTesseract
{
    Configuration = 
    {
        RenderSearchablePdf = true,
        PageSegmentationMode = TesseractPageSegmentationMode.Auto
    }
};

// Load a multi-page PDF
using var pdfInput = new OcrPdfInput("multi-page-scan.pdf");

// Optionally specify page range (e.g., pages 1-10)
pdfInput.SelectPages(1, 10);

// Perform OCR with progress tracking
OcrResult result = ocrTesseract.Read(pdfInput);

// Save as searchable PDF
result.SaveAsSearchablePdf("searchable-multi-page.pdf");

// Display total pages processed
Console.WriteLine($"Processed {result.Pages.Length} pages");

How Can I Apply Filters When Creating Searchable PDFs?

The SaveAsSearchablePdf also accepts a boolean flag as a second parameter that allows you to apply filters to a searchable PDF or not, giving developers the flexibility to choose. Using image optimization filters can significantly improve OCR accuracy, especially when dealing with low-quality scans.

Below is an example of applying the grayscale filter and then saving the PDF with a filter by putting true in the second parameter of SaveAsSearchablePdf.

:path=/static-assets/ocr/content-code-examples/how-to/image-quality-correction-searchable-pdf.cs

using IronOcr;

var ocr = new IronTesseract();
var ocrInput = new OcrInput();

// Load a PDF file
ocrInput.LoadPdf("invoice.pdf");

// Apply gray scale filter
ocrInput.ToGrayScale();
OcrResult result = ocr.Read(ocrInput);

// Save the result as a searchable PDF with filters applied
result.SaveAsSearchablePdf("outputGrayscale.pdf", true);

For optimal results, consider using the Filter Wizard to automatically determine the best combination of filters for your specific document type. This tool analyzes your input and suggests appropriate preprocessing steps.

Performance Considerations

When processing large volumes of documents, consider implementing multithreaded OCR operations to improve throughput. IronOCR supports concurrent processing, allowing you to handle multiple documents simultaneously:

using IronOcr;
using System.Threading.Tasks;
using System.Collections.Concurrent;

public class BatchPdfProcessor
{
    private readonly IronTesseract _ocr;

    public BatchPdfProcessor()
    {
        _ocr = new IronTesseract
        {
            Configuration = 
            {
                RenderSearchablePdf = true,
                // Configure for optimal performance
                Language = OcrLanguage.English
            }
        };
    }

    public async Task ProcessBatchAsync(string[] filePaths)
    {
        var results = new ConcurrentBag<(string source, string output)>();

        await Parallel.ForEachAsync(filePaths, async (filePath, ct) =>
        {
            using var input = new OcrImageInput(filePath);
            var result = _ocr.Read(input);

            string outputPath = Path.ChangeExtension(filePath, ".searchable.pdf");
            result.SaveAsSearchablePdf(outputPath);

            results.Add((filePath, outputPath));
        });

        Console.WriteLine($"Processed {results.Count} files");
    }
}

using IronOcr;
using System.Threading.Tasks;
using System.Collections.Concurrent;

public class BatchPdfProcessor
{
    private readonly IronTesseract _ocr;

    public BatchPdfProcessor()
    {
        _ocr = new IronTesseract
        {
            Configuration = 
            {
                RenderSearchablePdf = true,
                // Configure for optimal performance
                Language = OcrLanguage.English
            }
        };
    }

    public async Task ProcessBatchAsync(string[] filePaths)
    {
        var results = new ConcurrentBag<(string source, string output)>();

        await Parallel.ForEachAsync(filePaths, async (filePath, ct) =>
        {
            using var input = new OcrImageInput(filePath);
            var result = _ocr.Read(input);

            string outputPath = Path.ChangeExtension(filePath, ".searchable.pdf");
            result.SaveAsSearchablePdf(outputPath);

            results.Add((filePath, outputPath));
        });

        Console.WriteLine($"Processed {results.Count} files");
    }
}

Advanced Configuration Options

For more advanced scenarios, you can leverage detailed Tesseract configuration to fine-tune the OCR engine for specific document types or languages:

var advancedOcr = new IronTesseract
{
    Configuration = 
    {
        RenderSearchablePdf = true,
        TesseractVariables = new Dictionary<string, object>
        {
            { "preserve_interword_spaces", 1 },
            { "tessedit_char_whitelist", "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz" }
        },
        PageSegmentationMode = TesseractPageSegmentationMode.SingleColumn
    },
    Language = OcrLanguage.EnglishBest
};

var advancedOcr = new IronTesseract
{
    Configuration = 
    {
        RenderSearchablePdf = true,
        TesseractVariables = new Dictionary<string, object>
        {
            { "preserve_interword_spaces", 1 },
            { "tessedit_char_whitelist", "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz" }
        },
        PageSegmentationMode = TesseractPageSegmentationMode.SingleColumn
    },
    Language = OcrLanguage.EnglishBest
};