Update ocr_engine.py

ocr_engine.py

@@ -1,30 +1,41 @@
 import pytesseract
 import numpy as np
-import cv2
 import re
+import cv2
+from PIL import Image
 
 def extract_weight_from_image(pil_img):
     try:
+        # Convert PIL image to numpy array
         img = np.array(pil_img)
 
         # Convert to grayscale
         gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
 
-        # Resize for better OCR
-        gray = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
+        # Resize image to improve OCR accuracy
+        resized = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
+
+        # Apply Gaussian blur
+        blurred = cv2.GaussianBlur(resized, (5, 5), 0)
 
-        # Thresholding
-        _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
+        # Apply adaptive thresholding for better contrast
+        thresh = cv2.adaptiveThreshold(
+            blurred, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY_INV, 11, 2
+        )
 
-        # OCR using pytesseract
-        text = pytesseract.image_to_string(thresh, config='--psm 6 digits')
-        print("OCR Output:", text)
+        # OCR config: use digit-only mode and whitelist characters
+        config = "--psm 7 -c tessedit_char_whitelist=0123456789."
 
-        # Regex to extract weight (like 54.20, 102.5, etc.)
-        match = re.search(r"\b\d{2,4}\.?\d{0,2}\b", text)
+        # Extract text using pytesseract
+        ocr_text = pytesseract.image_to_string(thresh, config=config)
+        print("OCR Text:", ocr_text)
+
+        # Use regex to find weight values (e.g., 52.35, 002.50 etc.)
+        match = re.search(r"\b\d{2,4}\.?\d{0,2}\b", ocr_text)
         if match:
             return match.group(), 95.0
         else:
             return "No weight detected", 0.0
+
     except Exception as e:
         return f"Error: {str(e)}", 0.0