在 .NET 中使用 Tesseract 识别图片文字

dotnet add package Tesseract

dotnet add package SkiaSharp

public class TesseractOCRService{ private readonly HttpClient _httpClient; private readonly ILogger<TesseractOCRService> _logger; public TesseractOCRService(HttpClient httpClient, ILogger<TesseractOCRService> logger) { _httpClient = httpClient; _logger = logger; } public async Task Do(string imageUrl) { try { byte[] imageBytes = await DownloadImageAsync(imageUrl); using (var skBitmap = SKBitmap.Decode(imageBytes)) { var processedBitmap = PreprocessImage(skBitmap); using (var pix = ConvertSKBitmapToPix(processedBitmap)) { string captchaText = RecognizeText(pix); return captchaText; } } } catch (Exception ex) { _logger.LogError(ex.ToString()); return string.Empty; } } private async Task DownloadImageAsync(string url) { return await _httpClient.GetByteArrayAsync(url); } private SKBitmap PreprocessImage(SKBitmap bitmap) { return bitmap; } private Pix ConvertSKBitmapToPix(SKBitmap skBitmap) { using (var image = SKImage.FromBitmap(skBitmap)) using (var data = image.Encode(SKEncodedImageFormat.Png, 100)) { return Pix.LoadFromMemory(data.ToArray()); } } private string RecognizeText(Pix pix) { using (var engine = new TesseractEngine(@"./Services/TesseractService", "eng", EngineMode.Default)) { using (var page = engine.Process(pix)) { return page.GetText(); } } }}

RUN apt-get update && apt-get install -y \ libfontconfig1 \ libfreetype6 \ libpng16-16 \ libjpeg62-turbo \ libx11-6 \ libxext6 \ libxrender1 \ tesseract-ocr \ libtesseract-dev \ libleptonica-devRUN ln -s /usr/lib/x86_64-linux-gnu/libdl.so.2 /usr/lib/x86_64-linux-gnu/libdl.soWORKDIR /app/x64RUN ln -s /usr/lib/x86_64-linux-gnu/liblept.so.5 /app/x64/libleptonica-1.82.0.soRUN ln -s /usr/lib/x86_64-linux-gnu/libtesseract.so.5 /app/x64/libtesseract50.so