Update ocr_engine.py

ocr_engine.py

@@ -1,27 +1,28 @@
-import easyocr
 import numpy as np
 import re
 import cv2
-
-reader = easyocr.Reader(['en'], gpu=False)
+import pytesseract
 
 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=2.5, fy=2.5, interpolation=cv2.INTER_CUBIC)
-        blurred = cv2.GaussianBlur(resized, (3, 3), 0)
-        _, thresh = cv2.threshold(blurred, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
 
-        # OCR
-        result = reader.readtext(thresh, detail=0)
-        combined_text = " ".join(result)
-        print("OCR Result:", combined_text)
+        # Resize (sharpens small digits)
+        gray = cv2.resize(gray, None, fx=2.5, fy=2.5, interpolation=cv2.INTER_CUBIC)
+
+        # Thresholding to clean up image
+        _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
+
+        # Run Tesseract OCR
+        custom_config = r'--oem 3 --psm 6'
+        text = pytesseract.image_to_string(thresh, config=custom_config)
+        print("OCR Text:", text)
 
-        # Improve regex to only match numbers with optional decimal
-        match = re.search(r"\b(?:\d{1,3}\.?\d{1,2}|\d{1,4})\b", combined_text)
+        # Extract weight pattern like 25.50 or 150
+        match = re.search(r"\b\d{1,4}\.?\d{0,2}\b", text)
         if match:
             return match.group(), 95.0
         else: