Update ocr_engine.py

ocr_engine.py

@@ -1,29 +1,23 @@
 import cv2
-import pytesseract
+import easyocr
 import re
 
-# Optional: manually set Tesseract path if needed
-# pytesseract.pytesseract.tesseract_cmd = r"/usr/bin/tesseract"
+reader = easyocr.Reader(['en'], gpu=False)
 
 def extract_weight_from_image(img_cv):
-    # Resize to enhance detail
+    # Resize to help with clarity
     img = cv2.resize(img_cv, None, fx=2.0, fy=2.0, interpolation=cv2.INTER_CUBIC)
 
     # Convert to grayscale if needed
     if len(img.shape) == 3:
         img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 
-    # Adaptive threshold to boost OCR contrast
-    img = cv2.adaptiveThreshold(img, 255, cv2.ADAPTIVE_THRESH_MEAN_C,
-                                cv2.THRESH_BINARY, 11, 3)
+    results = reader.readtext(img)
 
-    # OCR configuration to focus on digits and decimal points
-    config = "--psm 6 -c tessedit_char_whitelist=0123456789."
+    raw_text = " ".join([text for _, text, _ in results])
+    matches = re.findall(r'\d+\.\d+|\d+', raw_text)
 
-    text = pytesseract.image_to_string(img, config=config)
-
-    # Regex to extract decimal or integer weight
-    matches = re.findall(r'\d+\.\d+|\d+', text)
     weight = matches[0] if matches else "0.0"
-    confidence = 85 if matches else 0  # Simulated confidence
-    return weight, confidence, text
+    confidence = int(results[0][2] * 100) if results else 0
+
+    return weight, confidence, raw_text