Spire.PDF 11.7.14 が PDF から Excel への変換精度を向上する XlsxLineLayoutOptions.TextRecognizer に対応

PdfDocument doc = new PdfDocument();
doc.LoadFromFile("in.pdf");
XlsxLineLayoutOptions options = new XlsxLineLayoutOptions(false, false, false, true);
options.TextRecognizer = new TextRecognizer();
doc.ConvertOptions.SetPdfToXlsxOptions(options);
doc.SaveToFile("out.xlsx", Spire.Pdf.FileFormat.XLSX);
 
// niget install PaddleOCRSharp lib
using PaddleOCRSharp;
using Spire.Pdf.Conversion;

public class TextRecognizer : ITextRecognizer
{
    private static readonly PaddleOCREngine _engine;

    static TextRecognizer()
   

{         _engine = new PaddleOCREngine(null, “”);     }
    public string RecognizeGlyph(Stream glyphImageStream)

{         
       var image = new System.Drawing.Bitmap(glyphImageStream);
        // paint glyph in image center
        var fixImage = new System.Drawing.Bitmap(160, 240);
        using (Graphics g = Graphics.FromImage(fixImage))
       

{             g.DrawImage(image, new RectangleF(20, 20, fixImage.Width - 40, fixImage.Height - 40), new RectangleF(0, 0, image.Width, image.Height), GraphicsUnit.Pixel);         }
        var unicodeResult = _engine.DetectText(fixImage).Text;
        return unicodeResult;
}

}