finished ocr daemon update

tools/OcrDaemon/ImageUtils.cs

@@ -0,0 +1,89 @@
+namespace OcrDaemon;
+
+using System.Drawing;
+using System.Drawing.Imaging;
+using System.Runtime.InteropServices;
+using Tesseract;
+using SdImageFormat = System.Drawing.Imaging.ImageFormat;
+
+static class ImageUtils
+{
+    public static Pix BitmapToPix(Bitmap bitmap)
+    {
+        using var ms = new MemoryStream();
+        bitmap.Save(ms, SdImageFormat.Png);
+        return Pix.LoadFromMemory(ms.ToArray());
+    }
+
+    public static List<OcrLineResult> ExtractLinesFromPage(Page page, int offsetX, int offsetY)
+    {
+        var lines = new List<OcrLineResult>();
+        using var iter = page.GetIterator();
+        if (iter == null) return lines;
+
+        iter.Begin();
+
+        do
+        {
+            var words = new List<OcrWordResult>();
+            do
+            {
+                var wordText = iter.GetText(PageIteratorLevel.Word);
+                if (string.IsNullOrWhiteSpace(wordText)) continue;
+
+                float conf = iter.GetConfidence(PageIteratorLevel.Word);
+                if (conf < 50) continue; // reject low-confidence garbage from background bleed
+
+                if (iter.TryGetBoundingBox(PageIteratorLevel.Word, out var bounds))
+                {
+                    words.Add(new OcrWordResult
+                    {
+                        Text = wordText.Trim(),
+                        X = bounds.X1 + offsetX,
+                        Y = bounds.Y1 + offsetY,
+                        Width = bounds.Width,
+                        Height = bounds.Height,
+                    });
+                }
+            } while (iter.Next(PageIteratorLevel.TextLine, PageIteratorLevel.Word));
+
+            if (words.Count > 0)
+            {
+                var lineText = string.Join(" ", words.Select(w => w.Text));
+                lines.Add(new OcrLineResult { Text = lineText, Words = words });
+            }
+        } while (iter.Next(PageIteratorLevel.Block, PageIteratorLevel.TextLine));
+
+        return lines;
+    }
+
+    public static (byte[] gray, byte[] argb, int stride) BitmapToGrayAndArgb(Bitmap bmp)
+    {
+        int w = bmp.Width, h = bmp.Height;
+        var data = bmp.LockBits(new Rectangle(0, 0, w, h), ImageLockMode.ReadOnly, PixelFormat.Format32bppArgb);
+        byte[] argb = new byte[data.Stride * h];
+        Marshal.Copy(data.Scan0, argb, 0, argb.Length);
+        bmp.UnlockBits(data);
+        int stride = data.Stride;
+
+        byte[] gray = new byte[w * h];
+        for (int y = 0; y < h; y++)
+            for (int x = 0; x < w; x++)
+            {
+                int i = y * stride + x * 4;
+                gray[y * w + x] = (byte)((argb[i] + argb[i + 1] + argb[i + 2]) / 3);
+            }
+        return (gray, argb, stride);
+    }
+
+    public static SdImageFormat GetImageFormat(string path)
+    {
+        var ext = Path.GetExtension(path).ToLowerInvariant();
+        return ext switch
+        {
+            ".jpg" or ".jpeg" => SdImageFormat.Jpeg,
+            ".bmp" => SdImageFormat.Bmp,
+            _ => SdImageFormat.Png,
+        };
+    }
+}