Update ocr_engine.py

ocr_engine.py

@@ -1,32 +1,48 @@
-import easyocr
+import numpy as np
 import re
 import cv2
-import numpy as np
 from PIL import Image
+import easyocr
+import os
 
-# Initialize EasyOCR reader
+# Initialize OCR Reader
 reader = easyocr.Reader(['en'], gpu=False)
 
 def preprocess_image(image):
+    """Preprocess the image to improve OCR accuracy"""
     # Convert to grayscale
     gray = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY)
 
-    # Blur + Otsu thresholding
-    blur = cv2.GaussianBlur(gray, (3, 3), 0)
-    _, thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
+    # Apply threshold to isolate digits
+    _, thresh = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)
 
     return thresh
 
 def extract_weight_from_image(pil_image):
     try:
-        # Convert PIL image to OpenCV image
+        # Convert PIL to OpenCV format
         image = np.array(pil_image.convert("RGB"))
+
+        # Print image shape for debugging
+        print("Image shape:", image.shape)
+
+        # Preprocess for better OCR accuracy
         processed = preprocess_image(image)
 
-        # OCR
+        # Save debug image
+        debug_img = Image.fromarray(processed)
+        debug_path = "debug_processed_image.png"
+        debug_img.save(debug_path)
+        print(f"✅ Processed image saved to: {debug_path}")
+
+        # Run OCR on processed image
         result = reader.readtext(processed)
-        print("OCR Results:", result)  # for debugging
 
+        print("✅ OCR Results:")
+        for r in result:
+            print(f"Text: '{r[1]}' | Confidence: {r[2] * 100:.2f}%")
+
+        # Try to find numeric weight
         weight = None
         confidence = 0.0
 
@@ -34,7 +50,8 @@ def extract_weight_from_image(pil_image):
             text = detection[1]
             conf = detection[2]
 
-            match = re.search(r"\b\d+(\.\d+)?\b", text)  # more flexible matching
+            # Match numbers like 53.25 or 45
+            match = re.search(r"\b\d+(\.\d+)?\b", text)
             if match:
                 weight = match.group()
                 confidence = conf
@@ -46,4 +63,5 @@ def extract_weight_from_image(pil_image):
             return "No weight detected", 0.0
 
     except Exception as e:
+        print("❌ Exception during OCR:", str(e))
         return f"Error: {str(e)}", 0.0