IronOCR 과 AWS Textract 비교: 스타트업을 위한 OCR 솔루션

using GdPicture14;

// Every image load returns an integer handle
int imageId = _imaging.CreateGdPictureImageFromFile(imagePath);

if (imageId == 0)
{
    throw new Exception($"Failed to load image: {_imaging.GetStat()}");
}

try
{
    _ocr.SetImage(imageId);
    _ocr.Language = "eng";

    string resultId = _ocr.RunOCR();

    if (string.IsNullOrEmpty(resultId))
    {
        throw new Exception($"OCR failed: {_ocr.GetStat()}");
    }

    return _ocr.GetOCRResultText(resultId);
}
finally
{
    // CRITICAL: omitting this line causes a memory leak
    _imaging.ReleaseGdPictureImage(imageId);
}

Imports GdPicture14

' Every image load returns an integer handle
Dim imageId As Integer = _imaging.CreateGdPictureImageFromFile(imagePath)

If imageId = 0 Then
    Throw New Exception($"Failed to load image: {_imaging.GetStat()}")
End If

Try
    _ocr.SetImage(imageId)
    _ocr.Language = "eng"

    Dim resultId As String = _ocr.RunOCR()

    If String.IsNullOrEmpty(resultId) Then
        Throw New Exception($"OCR failed: {_ocr.GetStat()}")
    End If

    Return _ocr.GetOCRResultText(resultId)
Finally
    ' CRITICAL: omitting this line causes a memory leak
    _imaging.ReleaseGdPictureImage(imageId)
End Try

using GdPicture14;

public string ExtractTextFromPdf(string pdfPath)
{
    var text = new StringBuilder();

    using var pdf = new GdPicturePDF();

    GdPictureStatus status = pdf.LoadFromFile(pdfPath, false);
    if (status != GdPictureStatus.OK)
        throw new Exception($"Failed to load PDF: {status}");

    int pageCount = pdf.GetPageCount();

    for (int i = 1; i <= pageCount; i++)
    {
        pdf.SelectPage(i);

        // Each page render = new memory allocation + new integer ID
        int imageId = pdf.RenderPageToGdPictureImage(200, false);

        if (imageId == 0) continue;

        try
        {
            _ocr.SetImage(imageId);
            _ocr.Language = "eng";
            string resultId = _ocr.RunOCR();

            if (!string.IsNullOrEmpty(resultId))
                text.AppendLine(_ocr.GetOCRResultText(resultId));
        }
        finally
        {
            // Without this: ~10–50MB leaked per page
            _imaging.ReleaseGdPictureImage(imageId);
        }
    }

    return text.ToString();
}

Imports GdPicture14
Imports System.Text

Public Function ExtractTextFromPdf(ByVal pdfPath As String) As String
    Dim text As New StringBuilder()

    Using pdf As New GdPicturePDF()
        Dim status As GdPictureStatus = pdf.LoadFromFile(pdfPath, False)
        If status <> GdPictureStatus.OK Then
            Throw New Exception($"Failed to load PDF: {status}")
        End If

        Dim pageCount As Integer = pdf.GetPageCount()

        For i As Integer = 1 To pageCount
            pdf.SelectPage(i)

            ' Each page render = new memory allocation + new integer ID
            Dim imageId As Integer = pdf.RenderPageToGdPictureImage(200, False)

            If imageId = 0 Then Continue For

            Try
                _ocr.SetImage(imageId)
                _ocr.Language = "eng"
                Dim resultId As String = _ocr.RunOCR()

                If Not String.IsNullOrEmpty(resultId) Then
                    text.AppendLine(_ocr.GetOCRResultText(resultId))
                End If
            Finally
                ' Without this: ~10–50MB leaked per page
                _imaging.ReleaseGdPictureImage(imageId)
            End Try
        Next
    End Using

    Return text.ToString()
End Function

using IronOcr;

public string ExtractTextFromPdf(string pdfPath)
{
    // LoadPdf handles page rendering internally
    using var input = new OcrInput();
    input.LoadPdf(pdfPath);

    var result = new IronTesseract().Read(input);

    foreach (var page in result.Pages)
        Console.WriteLine($"Page {page.PageNumber}: {page.Text}");

    return result.Text;
    // using block exits — all internal resources released automatically
}

Imports IronOcr

Public Function ExtractTextFromPdf(pdfPath As String) As String
    ' LoadPdf handles page rendering internally
    Using input As New OcrInput()
        input.LoadPdf(pdfPath)

        Dim result = New IronTesseract().Read(input)

        For Each page In result.Pages
            Console.WriteLine($"Page {page.PageNumber}: {page.Text}")
        Next

        Return result.Text
        ' using block exits — all internal resources released automatically
    End Using
End Function

using var input = new OcrInput();
input.LoadPdf("encrypted.pdf", Password: "secret");

var result = new IronTesseract().Read(input);
result.SaveAsSearchablePdf("searchable-output.pdf");

Imports IronOcr

Using input As New OcrInput()
    input.LoadPdf("encrypted.pdf", Password:="secret")

    Dim result = New IronTesseract().Read(input)
    result.SaveAsSearchablePdf("searchable-output.pdf")
End Using

using GdPicture14;

public void CreateSearchablePdf(string inputPdf, string outputPdf)
{
    using var pdf = new GdPicturePDF();

    GdPictureStatus status = pdf.LoadFromFile(inputPdf, false);
    if (status != GdPictureStatus.OK)
        throw new Exception($"Load failed: {status}");

    int pageCount = pdf.GetPageCount();

    for (int i = 1; i <= pageCount; i++)
    {
        pdf.SelectPage(i);

        // OcrPage adds text layer to each page
        // Requires: OCR plugin + PDF plugin + resource folder path
        GdPictureStatus ocrStatus = pdf.OcrPage(
            "eng",
            @"C:\GdPicture\Resources\OCR",  // 외부 리소스 폴더 required
            "",
            200
        );

        if (ocrStatus != GdPictureStatus.OK)
            Console.WriteLine($"Warning: Page {i} OCR failed: {ocrStatus}");
    }

    pdf.SaveToFile(outputPdf, true);
}

Imports GdPicture14

Public Sub CreateSearchablePdf(inputPdf As String, outputPdf As String)
    Using pdf As New GdPicturePDF()
        Dim status As GdPictureStatus = pdf.LoadFromFile(inputPdf, False)
        If status <> GdPictureStatus.OK Then
            Throw New Exception($"Load failed: {status}")
        End If

        Dim pageCount As Integer = pdf.GetPageCount()

        For i As Integer = 1 To pageCount
            pdf.SelectPage(i)

            ' OcrPage adds text layer to each page
            ' Requires: OCR plugin + PDF plugin + resource folder path
            Dim ocrStatus As GdPictureStatus = pdf.OcrPage(
                "eng",
                "C:\GdPicture\Resources\OCR",  ' 외부 리소스 폴더 required
                "",
                200
            )

            If ocrStatus <> GdPictureStatus.OK Then
                Console.WriteLine($"Warning: Page {i} OCR failed: {ocrStatus}")
            End If
        Next

        pdf.SaveToFile(outputPdf, True)
    End Using
End Sub

using IronOcr;

public void CreateSearchablePdf(string inputPdf, string outputPdf)
{
    var ocr = new IronTesseract();

    using var input = new OcrInput();
    input.LoadPdf(inputPdf);
    input.Deskew();   // 아니요 separate plugin required
    input.DeNoise();

    var result = ocr.Read(input);
    result.SaveAsSearchablePdf(outputPdf);
}

Imports IronOcr

Public Sub CreateSearchablePdf(inputPdf As String, outputPdf As String)
    Dim ocr = New IronTesseract()

    Using input = New OcrInput()
        input.LoadPdf(inputPdf)
        input.Deskew() ' 아니요 separate plugin required
        input.DeNoise()

        Dim result = ocr.Read(input)
        result.SaveAsSearchablePdf(outputPdf)
    End Using
End Sub

var result = new IronTesseract().Read("scanned.jpg");
Console.WriteLine($"Confidence: {result.Confidence}%");

// Per-word confidence for fine-grained filtering
var highConfidenceWords = result.Words
    .Where(w => w.Confidence > 85)
    .Select(w => w.Text);

Imports System
Imports System.Linq

Dim result = (New IronTesseract()).Read("scanned.jpg")
Console.WriteLine($"Confidence: {result.Confidence}%")

' Per-word confidence for fine-grained filtering
Dim highConfidenceWords = result.Words _
    .Where(Function(w) w.Confidence > 85) _
    .Select(Function(w) w.Text)

using GdPicture14;

public List<(string word, float confidence)> GetWordConfidences(int imageId)
{
    var words = new List<(string, float)>();

    _ocr.SetImage(imageId);
    _ocr.Language = "eng";
    string resultId = _ocr.RunOCR();

    int blockCount = _ocr.GetOCRResultBlockCount(resultId);

    for (int b = 0; b < blockCount; b++)
    {
        int lineCount = _ocr.GetOCRResultBlockLineCount(resultId, b);

        for (int l = 0; l < lineCount; l++)
        {
            int wordCount = _ocr.GetOCRResultBlockLineWordCount(resultId, b, l);

            for (int w = 0; w < wordCount; w++)
            {
                string word = _ocr.GetOCRResultBlockLineWordText(resultId, b, l, w);
                float conf = _ocr.GetOCRResultBlockLineWordConfidence(resultId, b, l, w);
                words.Add((word, conf));
            }
        }
    }

    return words;
}

Imports GdPicture14

Public Function GetWordConfidences(imageId As Integer) As List(Of (word As String, confidence As Single))
    Dim words As New List(Of (String, Single))()

    _ocr.SetImage(imageId)
    _ocr.Language = "eng"
    Dim resultId As String = _ocr.RunOCR()

    Dim blockCount As Integer = _ocr.GetOCRResultBlockCount(resultId)

    For b As Integer = 0 To blockCount - 1
        Dim lineCount As Integer = _ocr.GetOCRResultBlockLineCount(resultId, b)

        For l As Integer = 0 To lineCount - 1
            Dim wordCount As Integer = _ocr.GetOCRResultBlockLineWordCount(resultId, b, l)

            For w As Integer = 0 To wordCount - 1
                Dim word As String = _ocr.GetOCRResultBlockLineWordText(resultId, b, l, w)
                Dim conf As Single = _ocr.GetOCRResultBlockLineWordConfidence(resultId, b, l, w)
                words.Add((word, conf))
            Next
        Next
    Next

    Return words
End Function

using IronOcr;

var result = new IronTesseract().Read("invoice.jpg");

// All words with position and confidence
foreach (var word in result.Words)
{
    Console.WriteLine($"'{word.Text}' at ({word.X},{word.Y}) — {word.Confidence:P0}");
}

// Lines, paragraphs, and pages follow the same pattern
foreach (var line in result.Lines)
    Console.WriteLine(line.Text);

foreach (var page in result.Pages)
    Console.WriteLine($"Page {page.PageNumber}: {page.Text.Length} characters");

Imports IronOcr

Dim result = New IronTesseract().Read("invoice.jpg")

' All words with position and confidence
For Each word In result.Words
    Console.WriteLine($"'{word.Text}' at ({word.X},{word.Y}) — {word.Confidence:P0}")
Next

' Lines, paragraphs, and pages follow the same pattern
For Each line In result.Lines
    Console.WriteLine(line.Text)
Next

For Each page In result.Pages
    Console.WriteLine($"Page {page.PageNumber}: {page.Text.Length} characters")
Next

var region = new CropRectangle(0, 0, 600, 100); // header area

using var input = new OcrInput();
input.LoadImage("invoice.jpg", region);

var headerText = new IronTesseract().Read(input).Text;

Imports IronOcr

Dim region As New CropRectangle(0, 0, 600, 100) ' header area

Using input As New OcrInput()
    input.LoadImage("invoice.jpg", region)

    Dim headerText As String = New IronTesseract().Read(input).Text
End Using

// Remove
LicenseManager lm = new LicenseManager();
lm.RegisterKEY("GDPICTURE-LICENSE-KEY");
_imaging = new GdPictureImaging();
_ocr = new GdPictureOCR();
_ocr.ResourceFolder = @"C:\GdPicture\Resources\OCR";

// Replace with
IronOcr.License.LicenseKey = "IRONOCR-LICENSE-KEY";
// IronTesseract instances are instantiated per-use or as a singleton

' Remove
Dim lm As New LicenseManager()
lm.RegisterKEY("GDPICTURE-LICENSE-KEY")
_imaging = New GdPictureImaging()
_ocr = New GdPictureOCR()
_ocr.ResourceFolder = "C:\GdPicture\Resources\OCR"

' Replace with
IronOcr.License.LicenseKey = "IRONOCR-LICENSE-KEY"
' IronTesseract instances are instantiated per-use or as a singleton

// Remove
int imageId = _imaging.CreateGdPictureImageFromFile(path);
if (imageId == 0) throw new Exception(_imaging.GetStat().ToString());
_ocr.SetImage(imageId);
_ocr.Language = "eng";
string resultId = _ocr.RunOCR();
if (string.IsNullOrEmpty(resultId)) throw new Exception(_ocr.GetStat().ToString());
string text = _ocr.GetOCRResultText(resultId);
_imaging.ReleaseGdPictureImage(imageId);

// Replace with
using var input = new OcrInput();
input.LoadImage(path);
var result = new IronTesseract().Read(input);
string text = result.Text;

Imports System

' Remove
Dim imageId As Integer = _imaging.CreateGdPictureImageFromFile(path)
If imageId = 0 Then Throw New Exception(_imaging.GetStat().ToString())
_ocr.SetImage(imageId)
_ocr.Language = "eng"
Dim resultId As String = _ocr.RunOCR()
If String.IsNullOrEmpty(resultId) Then Throw New Exception(_ocr.GetStat().ToString())
Dim text As String = _ocr.GetOCRResultText(resultId)
_imaging.ReleaseGdPictureImage(imageId)

' Replace with
Using input As New OcrInput()
    input.LoadImage(path)
    Dim result = New IronTesseract().Read(input)
    Dim text As String = result.Text
End Using

# Remove filesystem dependency for each language
# Add NuGet package instead
dotnet add package IronOcr.Languages.French
dotnet add package IronOcr.Languages.German
dotnet add package IronOcr.Languages.ChineseSimplified