Update ocr_engine.py

ocr_engine.py

@@ -1,25 +1,34 @@
-import cv2
-import pytesseract
-import numpy as np
-from PIL import Image
+try:
+    from paddleocr import PaddleOCR
+    import re
+    ocr = PaddleOCR(use_angle_cls=True, lang='en')
+except Exception as e:
+    print(f"❌ PaddleOCR failed to load: {e}")
+    ocr = None
 
-def extract_weight_from_image(pil_img):
-    try:
-        # Convert PIL image to OpenCV
-        img = pil_img.convert("RGB")
-        img = np.array(img)
-        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+def extract_weight_from_image(image):
+    if ocr is None:
+        return ("OCR not initialized", 0.0)
 
-        # Preprocess
-        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-        blur = cv2.GaussianBlur(gray, (3, 3), 0)
+    result = ocr.ocr(image, cls=True)
+    debug_texts = []
 
-        # OCR
-        text = pytesseract.image_to_string(blur, config='--psm 7 digits')
-        weight = ''.join(filter(lambda c: c in '0123456789.', text))
+    if not result or not result[0]:
+        return ("No weight detected", 0.0)
 
-        confidence = 95  # Replace with real confidence logic if needed
-        return weight.strip(), confidence
-    except Exception as e:
-        print(f"OCR error: {e}")
-        return "", 0
+    for line in result[0]:
+        text, confidence = line[1][0], line[1][1]
+        debug_texts.append(f"{text} (Conf: {confidence:.2f})")
+
+        # Regex: number with optional kg/g
+        match = re.search(r'(\d+\.?\d*)\s*(kg|g)?', text.lower())
+        if match:
+            weight = match.group(1)
+            unit = match.group(2) if match.group(2) else "g"
+            return (f"{weight} {unit}", confidence)
+
+    print("🧪 OCR DEBUG:")
+    for t in debug_texts:
+        print(t)
+
+    return ("No weight detected", 0.0)