Changed OCR process

app/services/ocr_service.py

@@ -1,6 +1,7 @@
 import pytesseract
 import re
-from PIL import Image
+from PIL import Image, ImageFilter, ImageEnhance
+import numpy as np
 import sys
 import os
 
@@ -8,37 +9,39 @@ import os
 class OCRService:
 
     def __init__(self):
-        # Auto-detect tesseract path
         if sys.platform.startswith("win"):
             pytesseract.pytesseract.tesseract_cmd = os.getenv("TESSERACT_PATH", "C:/Program Files/Tesseract-OCR/tesseract.exe")
         else:
-            # Linux / Hugging Face Spaces
             pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"
-        print(f"OCR service initialized. Using Tesseract at {pytesseract.pytesseract.tesseract_cmd}")
+
+    def _preprocess(self, region: Image.Image, scale: int = 3) -> Image.Image:
+        """Upscale, convert to grayscale, and threshold for better OCR."""
+        region = region.resize(
+            (region.width * scale, region.height * scale),
+            Image.LANCZOS
+        )
+        region = region.convert("L")  # grayscale
+        # Increase contrast
+        region = ImageEnhance.Contrast(region).enhance(2.0)
+        # Threshold to black/white
+        region = region.point(lambda x: 0 if x < 140 else 255, "1").convert("L")
+        return region
 
     def extract(self, image: Image.Image) -> dict:
         w, h = image.size
 
-        # --------------------------------
-        # ----- CROP REGIONS -------------
-        # --------------------------------
-
-        # Card name — top left area
-        name_region = image.crop((0.15 * w, 0.02 * h, 0.75 * w, 0.10 * h))
+        # Name — skip "Basic Pokemon" line at very top, just grab name row
+        name_region = image.crop((0.05 * w, 0.06 * h, 0.72 * w, 0.13 * h))
 
-        # HP — top right area
-        hp_region = image.crop((0.60 * w, 0.02 * h, 0.95 * w, 0.10 * h))
+        # HP — top right, large number + "HP" text
+        hp_region = image.crop((0.55 * w, 0.04 * h, 0.97 * w, 0.13 * h))
 
-        # Moves — lower middle section
-        moves_region = image.crop((0.00 * w, 0.55 * h, 1.00 * w, 0.85 * h))
+        # Moves — middle to lower section
+        moves_region = image.crop((0.02 * w, 0.52 * h, 0.98 * w, 0.88 * h))
 
         # Full image for type detection
         full_text = pytesseract.image_to_string(image)
 
-        # --------------------------------
-        # ----- EXTRACT FIELDS -----------
-        # --------------------------------
-
         return {
             "name": self._extract_name(name_region),
             "hp": self._extract_hp(hp_region),
@@ -46,29 +49,27 @@ class OCRService:
             "moves": self._extract_moves(moves_region),
         }
 
-    # --------------------------------
-    # ----- EXTRACTORS ---------------
-    # --------------------------------
-
     def _extract_name(self, region: Image.Image) -> str | None:
-        # Upscale region for better OCR accuracy
-        region = region.resize(
-            (region.width * 3, region.height * 3),
-            Image.LANCZOS
-        )
-        text = pytesseract.image_to_string(region, config="--psm 7").strip()
+        region = self._preprocess(region, scale=3)
+        text = pytesseract.image_to_string(region, config="--psm 7 --oem 3").strip()
+        # Clean up noise — keep only lines that look like a name
+        lines = [l.strip() for l in text.splitlines() if l.strip()]
+        for line in lines:
+            # Skip lines that are clearly not a name
+            if re.search(r'[A-Z][a-z]+', line) and len(line) < 30:
+                return line
         return text if text else None
 
     def _extract_hp(self, region: Image.Image) -> str | None:
-        region = region.resize(
-            (region.width * 3, region.height * 3),
-            Image.LANCZOS
-        )
-        text = pytesseract.image_to_string(region, config="--psm 7")
+        region = self._preprocess(region, scale=3)
+        text = pytesseract.image_to_string(region, config="--psm 6 --oem 3")
+        # Look for a number near "HP"
         match = re.search(r'(\d+)\s*HP|HP\s*(\d+)', text, re.IGNORECASE)
         if match:
             return match.group(1) or match.group(2)
-        return None
+        # Fallback: just grab any standalone number (the HP value)
+        match = re.search(r'\b(\d{2,3})\b', text)
+        return match.group(1) if match else None
 
     def _extract_types(self, text: str) -> list[str] | None:
         types = [
@@ -80,25 +81,33 @@ class OCRService:
         return found if found else None
 
     def _extract_moves(self, region: Image.Image) -> list[dict] | None:
-        region = region.resize(
-            (region.width * 2, region.height * 2),
-            Image.LANCZOS
-        )
-        text = pytesseract.image_to_string(region)
+        region = self._preprocess(region, scale=2)
+        text = pytesseract.image_to_string(region, config="--psm 6 --oem 3")
         lines = [line.strip() for line in text.splitlines() if line.strip()]
 
         moves = []
         i = 0
         while i < len(lines):
-            # Match move name with damage e.g. "Lightning Flash 20"
-            match = re.match(r'^([A-Z][a-zA-Z\s]+?)\s+(\d+\+?)$', lines[i])
+            # Match: "MoveName 10" or "MoveName 10+" or "MoveName" alone on a line (0 damage moves)
+            match = re.match(r'^([A-Z][a-zA-Z\s]{2,25}?)\s{2,}(\d+\+?)$', lines[i])
+            if not match:
+                # Try looser match for lines like "Psychic                    10+"
+                match = re.match(r'^([A-Z][a-zA-Z]+)\s+(\d+\+?)$', lines[i])
+
             if match:
+                # Collect any following lines as move description until next move or end
+                desc_lines = []
+                j = i + 1
+                while j < len(lines) and not re.match(r'^[A-Z][a-zA-Z\s]+\s+\d+', lines[j]):
+                    desc_lines.append(lines[j])
+                    j += 1
+
                 moves.append({
                     "name": match.group(1).strip(),
                     "damage": match.group(2).strip(),
-                    "text": lines[i + 1] if i + 1 < len(lines) else None
+                    "text": " ".join(desc_lines) if desc_lines else None
                 })
-                i += 2
+                i = j
             else:
                 i += 1