IronOCR 与 Azure OCR PDF：哪种解决方案能更好地提取文本？

// MauiProgram.cs — this registration only compiles inside a MAUI project
public static MauiApp CreateMauiApp()
{
    var builder = MauiApp.CreateBuilder();
    builder.UseMauiApp<App>();

    // Registers ITesseract — MAUI DI only, no standalone equivalent
    builder.Services.AddTesseractOcr();

    return builder.Build();
}

// OCR service — ITesseract injected from MAUI container
public class MobileOcrService
{
    private readonly ITesseract _tesseract;

    public MobileOcrService(ITesseract tesseract)
    {
        _tesseract = tesseract; // Cannot be constructed outside MAUI host
    }

    public async Task<string> ExtractTextAsync(string imagePath)
    {
        await _tesseract.InitAsync("eng"); // traineddata must be bundled
        var result = await _tesseract.RecognizeTextAsync(imagePath);
        return result.Success ? result.RecognizedText : string.Empty;
    }
}

Imports Microsoft.Maui
Imports Microsoft.Extensions.DependencyInjection

' MauiProgram.vb — this registration only compiles inside a MAUI project
Public Module MauiProgram
    Public Function CreateMauiApp() As MauiApp
        Dim builder = MauiApp.CreateBuilder()
        builder.UseMauiApp(Of App)()

        ' Registers ITesseract — MAUI DI only, no standalone equivalent
        builder.Services.AddTesseractOcr()

        Return builder.Build()
    End Function
End Module

' OCR service — ITesseract injected from MAUI container
Public Class MobileOcrService
    Private ReadOnly _tesseract As ITesseract

    Public Sub New(tesseract As ITesseract)
        _tesseract = tesseract ' Cannot be constructed outside MAUI host
    End Sub

    Public Async Function ExtractTextAsync(imagePath As String) As Task(Of String)
        Await _tesseract.InitAsync("eng") ' traineddata must be bundled
        Dim result = Await _tesseract.RecognizeTextAsync(imagePath)
        Return If(result.Success, result.RecognizedText, String.Empty)
    End Function
End Class

// MAUI-only: will not compile in ASP.NET Core, console, or class library
public class DocumentScanService
{
    private readonly ITesseract _tesseract;

    public DocumentScanService(ITesseract tesseract) // MAUI DI injection only
    {
        _tesseract = tesseract;
    }

    public async Task<string> ProcessAsync(string imagePath)
    {
        await _tesseract.InitAsync("eng");
        var result = await _tesseract.RecognizeTextAsync(imagePath);
        return result.RecognizedText ?? string.Empty;
    }
}

Imports System.Threading.Tasks

' MAUI-only: will not compile in ASP.NET Core, console, or class library
Public Class DocumentScanService
    Private ReadOnly _tesseract As ITesseract

    Public Sub New(tesseract As ITesseract) ' MAUI DI injection only
        _tesseract = tesseract
    End Sub

    Public Async Function ProcessAsync(imagePath As String) As Task(Of String)
        Await _tesseract.InitAsync("eng")
        Dim result = Await _tesseract.RecognizeTextAsync(imagePath)
        Return If(result.RecognizedText, String.Empty)
    End Function
End Class

// Works in every project type — MAUI, ASP.NET Core, console, Azure Functions, Docker
public class SharedOcrService
{
    private readonly IronTesseract _ocr = new IronTesseract();

    public string ExtractText(string imagePath)
    {
        using var input = new OcrInput(imagePath);
        return _ocr.Read(input).Text;
    }

    public string ExtractTextFromBytes(byte[] imageData)
    {
        using var input = new OcrInput(imageData);
        return _ocr.Read(input).Text;
    }
}

Imports IronTesseract

Public Class SharedOcrService
    Private ReadOnly _ocr As New IronTesseract()

    Public Function ExtractText(imagePath As String) As String
        Using input As New OcrInput(imagePath)
            Return _ocr.Read(input).Text
        End Using
    End Function

    Public Function ExtractTextFromBytes(imageData As Byte()) As String
        Using input As New OcrInput(imageData)
            Return _ocr.Read(input).Text
        End Using
    End Function
End Class

// TesseractOcrMaui: PDF processing throws NotSupportedException
public async Task<List<string>> ProcessPdfAsync(string pdfPath)
{
    // Cannot process PDFs at all — requires manual conversion to images first
    // AND requires a separate PDF library you must add yourself
    throw new NotSupportedException(
        "TesseractOcrMaui does not support PDF processing. " +
        "You must manually convert PDF pages to images first.");
}

Imports System
Imports System.Threading.Tasks

Public Class TesseractOcrMaui
    Public Async Function ProcessPdfAsync(pdfPath As String) As Task(Of List(Of String))
        ' Cannot process PDFs at all — requires manual conversion to images first
        ' AND requires a separate PDF library you must add yourself
        Throw New NotSupportedException("TesseractOcrMaui does not support PDF processing. " &
                                        "You must manually convert PDF pages to images first.")
    End Function
End Class

//ASP.NET Corecontroller — handles PDF uploads directly
[HttpPost("ocr")]
public async Task<IActionResult> ProcessUpload(IFormFile file)
{
    var ocr = new IronTesseract();

    using var input = new OcrInput();

    if (file.ContentType == "application/pdf")
    {
        using var stream = file.OpenReadStream();
        input.LoadPdf(stream); // direct PDF stream — no temp files
    }
    else
    {
        using var stream = file.OpenReadStream();
        input.LoadImage(stream);
    }

    input.Deskew();   // handle scanned document skew
    input.DeNoise();  // mobile camera noise
    var result = ocr.Read(input);
    return Ok(new { text = result.Text, confidence = result.Confidence });
}

Imports Microsoft.AspNetCore.Mvc
Imports IronOcr

<HttpPost("ocr")>
Public Async Function ProcessUpload(file As IFormFile) As Task(Of IActionResult)
    Dim ocr As New IronTesseract()

    Using input As New OcrInput()
        If file.ContentType = "application/pdf" Then
            Using stream = file.OpenReadStream()
                input.LoadPdf(stream) ' direct PDF stream — no temp files
            End Using
        Else
            Using stream = file.OpenReadStream()
                input.LoadImage(stream)
            End Using
        End If

        input.Deskew() ' handle scanned document skew
        input.DeNoise() ' mobile camera noise
        Dim result = ocr.Read(input)
        Return Ok(New With {Key .text = result.Text, Key .confidence = result.Confidence})
    End Using
End Function

// Required startup validation — common crash source
public async Task<bool> ValidateTraineddataAsync()
{
    try
    {
        // Fails if eng.traineddata is missing, path-resolved incorrectly,
        // or mismatched with the Tesseract engine version
        await _tesseract.InitAsync("eng");
        return true;
    }
    catch (FileNotFoundException)
    {
        // Need to diagnose: missing file, wrong path, or wrong bundling config?
        Console.WriteLine("ERROR: eng.traineddata not found in bundle");
        return false;
    }
    catch (DllNotFoundException)
    {
        // Native library resolution failure — platform-specific debugging
        Console.WriteLine("Native Tesseract library missing");
        return false;
    }
}

Imports System
Imports System.IO
Imports System.Threading.Tasks

Public Class Validator
    ' Required startup validation — common crash source
    Public Async Function ValidateTraineddataAsync() As Task(Of Boolean)
        Try
            ' Fails if eng.traineddata is missing, path-resolved incorrectly,
            ' or mismatched with the Tesseract engine version
            Await _tesseract.InitAsync("eng")
            Return True
        Catch ex As FileNotFoundException
            ' Need to diagnose: missing file, wrong path, or wrong bundling config?
            Console.WriteLine("ERROR: eng.traineddata not found in bundle")
            Return False
        Catch ex As DllNotFoundException
            ' Native library resolution failure — platform-specific debugging
            Console.WriteLine("Native Tesseract library missing")
            Return False
        End Try
    End Function

    Private _tesseract As Tesseract ' Assuming Tesseract is a defined class elsewhere
End Class

// Application startup — the entire configuration
IronOcr.License.LicenseKey = "YOUR-LICENSE-KEY";

// Then OCR works anywhere, any project type, any file format
var text = new IronTesseract().Read("document.jpg").Text;

Imports IronOcr

' Application startup — the entire configuration
IronOcr.License.LicenseKey = "YOUR-LICENSE-KEY"

' Then OCR works anywhere, any project type, any file format
Dim text As String = New IronTesseract().Read("document.jpg").Text

// Before — MAUI DI required
public class OcrService
{
    private readonly ITesseract _tesseract;
    public OcrService(ITesseract tesseract) { _tesseract = tesseract; }

    public async Task<string> ProcessAsync(string imagePath)
    {
        await _tesseract.InitAsync("eng");
        var result = await _tesseract.RecognizeTextAsync(imagePath);
        return result.Success ? result.RecognizedText : string.Empty;
    }
}

// After — works in any project type
public class OcrService
{
    public string Process(string imagePath)
    {
        var ocr = new IronTesseract();
        using var input = new OcrInput(imagePath);
        return ocr.Read(input).Text;
    }
}

Imports System.Threading.Tasks

' Before — MAUI DI required
Public Class OcrService
    Private ReadOnly _tesseract As ITesseract

    Public Sub New(tesseract As ITesseract)
        _tesseract = tesseract
    End Sub

    Public Async Function ProcessAsync(imagePath As String) As Task(Of String)
        Await _tesseract.InitAsync("eng")
        Dim result = Await _tesseract.RecognizeTextAsync(imagePath)
        Return If(result.Success, result.RecognizedText, String.Empty)
    End Function
End Class

' After — works in any project type
Public Class OcrService
    Public Function Process(imagePath As String) As String
        Dim ocr = New IronTesseract()
        Using input = New OcrInput(imagePath)
            Return ocr.Read(input).Text
        End Using
    End Function
End Class

dotnet remove package TesseractOcrMaui

dotnet add package IronOcr
dotnet add package IronOcr.Android   # for MAUI Android target
dotnet add package IronOcr.iOS       # for MAUI iOS target
dotnet add package IronOcr.Languages.French  # if French was needed

using var input = new OcrInput(imagePath);
input.Deskew();            // camera angle correction
input.DeNoise();           // sensor noise from low light
input.EnhanceResolution(300); // normalize DPI across device models

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

Imports IronOcr

Using input As New OcrInput(imagePath)
    input.Deskew()            ' camera angle correction
    input.DeNoise()           ' sensor noise from low light
    input.EnhanceResolution(300) ' normalize DPI across device models

    Dim result = New IronTesseract().Read(input)
    Console.WriteLine($"Confidence: {result.Confidence}%")
End Using