Create ocr_engine

ocr_engine

@@ -0,0 +1,71 @@
+import easyocr
+import numpy as np
+import cv2
+import re
+
+# Load EasyOCR reader
+reader = easyocr.Reader(['en'], gpu=False)
+
+def extract_weight_from_image(pil_img):
+    try:
+        img = np.array(pil_img)
+
+        # Resize very large images
+        max_dim = 1000
+        height, width = img.shape[:2]
+        if max(height, width) > max_dim:
+            scale = max_dim / max(height, width)
+            img = cv2.resize(img, None, fx=scale, fy=scale, interpolation=cv2.INTER_AREA)
+
+        # OCR recognition
+        results = reader.readtext(img)
+        print("DEBUG OCR RESULTS:", results)
+
+        raw_texts = []
+        weight_candidates = []
+        fallback_weight = None
+        fallback_conf = 0.0
+
+        for _, (text, conf) in results:
+            original = text
+            cleaned = text.lower().strip()
+
+            # Fix common OCR misreads
+            cleaned = cleaned.replace(",", ".")
+            cleaned = cleaned.replace("o", "0").replace("O", "0")
+            cleaned = cleaned.replace("s", "5").replace("S", "5")
+            cleaned = cleaned.replace("g", "9").replace("G", "6")
+            cleaned = cleaned.replace("kg", "").replace("kgs", "")
+            cleaned = re.sub(r"[^0-9\.]", "", cleaned)
+
+            raw_texts.append(f"{original} → {cleaned} (conf: {round(conf, 2)})")
+
+            # Save fallback if no match later
+            if cleaned and cleaned.replace(".", "").isdigit() and not fallback_weight:
+                fallback_weight = cleaned
+                fallback_conf = conf
+
+            # Match proper weight format: 75.02, 97.2, 105
+            if cleaned.count(".") <= 1 and re.fullmatch(r"\d{2,4}(\.\d{1,3})?", cleaned):
+                weight_candidates.append((cleaned, conf))
+
+        # Choose best candidate
+        if weight_candidates:
+            best_weight, best_conf = sorted(weight_candidates, key=lambda x: -x[1])[0]
+        elif fallback_weight:
+            best_weight, best_conf = fallback_weight, fallback_conf
+        else:
+            return "Not detected", 0.0, "\n".join(raw_texts)
+
+        # Strip unnecessary leading zeros
+        if "." in best_weight:
+            int_part, dec_part = best_weight.split(".")
+            int_part = int_part.lstrip("0") or "0"
+            best_weight = f"{int_part}.{dec_part}"
+        else:
+            best_weight = best_weight.lstrip("0") or "0"
+
+        return best_weight, round(best_conf * 100, 2), "\n".join(raw_texts)
+
+    except Exception as e:
+        return f"Error: {str(e)}", 0.0, "OCR failed"