Update ocr_engine.py

ocr_engine.py

@@ -5,60 +5,34 @@ import re
 import logging
 from PIL import Image
 
-# Setup logging
 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
-def preprocess_strong(img):
-    """Sharpen and enhance contrast for blurry weight images."""
+def preprocess_image(img):
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-
-    # Resize up for OCR (scale x2)
-    h, w = gray.shape
-    gray = cv2.resize(gray, (w * 2, h * 2), interpolation=cv2.INTER_CUBIC)
-
-    # CLAHE for contrast enhancement
-    clahe = cv2.createCLAHE(clipLimit=3.0, tileGridSize=(8, 8))
-    enhanced = clahe.apply(gray)
-
-    # Strong sharpening
-    kernel = np.array([[0, -1, 0],
-                       [-1, 5, -1],
-                       [0, -1, 0]])
-    sharpened = cv2.filter2D(enhanced, -1, kernel)
-
-    return sharpened
+    resized = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_LINEAR)
+    blurred = cv2.GaussianBlur(resized, (3, 3), 0)
+    thresh = cv2.adaptiveThreshold(blurred, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
+                                   cv2.THRESH_BINARY_INV, 11, 2)
+    return thresh
 
 def extract_weight_from_image(pil_img):
-    """Extract weight from an image using multiple Tesseract strategies."""
     try:
         img = np.array(pil_img)
         img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
-
-        processed = preprocess_strong(img)
-
-        # OCR configs
-        configs = [
-            r'--oem 3 --psm 6 -c tessedit_char_whitelist=0123456789.',
-            r'--oem 3 --psm 11 -c tessedit_char_whitelist=0123456789.'
-        ]
-
-        for config in configs:
-            raw_text = pytesseract.image_to_string(processed, config=config)
-            logging.info(f"[Tesseract Output {config}] Raw text: {raw_text}")
-            cleaned = raw_text.strip().replace('\n', '').replace(' ', '')
-            cleaned = re.sub(r"[^\d.]", "", cleaned)
-            if cleaned.count('.') > 1:
-                cleaned = cleaned.replace('.', '', cleaned.count('.') - 1)
-            if cleaned.startswith('.'):
-                cleaned = '0' + cleaned
-
-            if cleaned and re.fullmatch(r"\d*\.?\d*", cleaned):
-                value = float(cleaned)
-                if 0.001 <= value <= 5000:
-                    return str(round(value, 2)), 90.0
-
+        processed = preprocess_image(img)
+
+        config = r'--oem 3 --psm 6 -c tessedit_char_whitelist=0123456789.kg'
+        raw_text = pytesseract.image_to_string(processed, config=config)
+        logging.info(f"OCR Raw Output: {raw_text}")
+
+        cleaned = raw_text.replace(" ", "").replace("\n", "")
+        match = re.search(r"(\d+\.?\d*)", cleaned)
+        if match:
+            value = float(match.group(1))
+            if 0 < value <= 5000:
+                return str(value), 90.0
         return "Not detected", 0.0
 
     except Exception as e:
-        logging.error(f"OCR failed: {e}")
+        logging.error(f"OCR error: {e}")
         return "Not detected", 0.0