Update ocr_engine.py

ocr_engine.py

@@ -7,19 +7,21 @@ def extract_weight_from_image(pil_img):
     try:
         img = np.array(pil_img)
 
-        # Preprocessing
+        # Convert to grayscale
         gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
-        resized = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
-        blur = cv2.GaussianBlur(resized, (5, 5), 0)
-        _, thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
+
+        # Resize for better OCR
+        gray = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
+
+        # Thresholding
+        _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
 
         # OCR using pytesseract
-        config = "--psm 6"  # Assume a single uniform block of text
-        text = pytesseract.image_to_string(thresh, config=config)
+        text = pytesseract.image_to_string(thresh, config='--psm 6 digits')
         print("OCR Output:", text)
 
-        # Regex to find weight-like numbers
-        match = re.search(r"\b\d{1,4}\.?\d{0,2}\b", text)
+        # Regex to extract weight (like 54.20, 102.5, etc.)
+        match = re.search(r"\b\d{2,4}\.?\d{0,2}\b", text)
         if match:
             return match.group(), 95.0
         else: