Update ocr_engine.py

ocr_engine.py

@@ -1,37 +1,31 @@
-import pytesseract
+import easyocr
 import numpy as np
 import re
 import cv2
-from PIL import Image
+
+reader = easyocr.Reader(['en'], gpu=False)
 
 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 image to improve OCR accuracy
+        # Resize to improve OCR accuracy
         resized = cv2.resize(gray, None, fx=2, fy=2, interpolation=cv2.INTER_CUBIC)
 
-        # Apply Gaussian blur
+        # Blur and threshold
         blurred = cv2.GaussianBlur(resized, (5, 5), 0)
+        _, thresh = cv2.threshold(blurred, 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 config: use digit-only mode and whitelist characters
-        config = "--psm 7 -c tessedit_char_whitelist=0123456789."
-
-        # Extract text using pytesseract
-        ocr_text = pytesseract.image_to_string(thresh, config=config)
-        print("OCR Text:", ocr_text)
+        # OCR
+        result = reader.readtext(thresh, detail=0)
+        text = " ".join(result)
+        print("OCR Text:", 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)
+        # Regex to find weight like 002.50 or 55.3
+        match = re.search(r"\b\d{2,4}\.?\d{0,2}\b", text)
         if match:
             return match.group(), 95.0
         else: