Update ocr_engine.py

ocr_engine.py

@@ -6,7 +6,6 @@ import re
 reader = easyocr.Reader(['en'], gpu=False)
 
 def enhance_image(img):
-    # Resize big images
     max_dim = 1000
     height, width = img.shape[:2]
     if max(height, width) > max_dim:
@@ -20,9 +19,9 @@ def enhance_image(img):
     sharp = cv2.filter2D(gray, -1, kernel)
 
     clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
-    contrast = clahe.apply(sharp)
+    enhanced = clahe.apply(sharp)
 
-    return contrast
+    return enhanced
 
 def extract_weight_from_image(pil_img):
     try:
@@ -32,34 +31,40 @@ def extract_weight_from_image(pil_img):
         results = reader.readtext(enhanced)
         print("DEBUG OCR RESULTS:", results)
 
-        ocr_texts = [text for _, text, _ in results]
+        if not results:
+            return "No text detected", 0.0, "OCR returned empty list"
+
+        all_texts = []
         weight_candidates = []
 
         for _, text, conf in results:
+            original = text
             cleaned = text.lower().strip()
-
-            # Fix common misreads
-            cleaned = cleaned.replace(",", ".")  # comma → dot
+            cleaned = cleaned.replace(",", ".")
             cleaned = cleaned.replace("o", "0").replace("O", "0")
             cleaned = cleaned.replace("s", "5").replace("S", "5")
             cleaned = cleaned.replace("g", "9").replace("G", "6")
             cleaned = cleaned.replace("kg", "").replace("kgs", "")
-            cleaned = re.sub(r"[^\d\.]", "", cleaned)  # Keep only digits + dot
+            cleaned = re.sub(r"[^\d\.]", "", cleaned)
+
+            all_texts.append(f"{original} → {cleaned} (conf: {round(conf, 2)})")
 
-            # Match: 2 to 4 digits, optional .digit
-            if re.fullmatch(r"\d{2,4}(\.\d{1,2})?", cleaned):
+            if re.fullmatch(r"\d{2,4}(\.\d{1,3})?", cleaned):
                 weight_candidates.append((cleaned, conf))
 
         if not weight_candidates:
-            return "Not detected", 0.0, "\n".join(ocr_texts)
+            return "Not detected", 0.0, "\n".join(all_texts)
 
-        # Get highest confidence result
         best_weight, best_conf = sorted(weight_candidates, key=lambda x: -x[1])[0]
 
-        # Remove leading zeros
-        best_weight = best_weight.lstrip('0') or '0'
+        if "." in best_weight:
+            parts = best_weight.split(".")
+            parts[0] = parts[0].lstrip("0") or "0"
+            best_weight = ".".join(parts)
+        else:
+            best_weight = best_weight.lstrip("0") or "0"
 
-        return best_weight, round(best_conf * 100, 2), "\n".join(ocr_texts)
+        return best_weight, round(best_conf * 100, 2), "\n".join(all_texts)
 
     except Exception as e:
         return f"Error: {str(e)}", 0.0, "OCR failed"