ocr_engine.py

import easyocr
import numpy as np
import cv2
import re

# Load the OCR engine
reader = easyocr.Reader(['en'], gpu=False)

def extract_weight_from_image(pil_img):
    try:
        # Convert PIL to OpenCV image (numpy array)
        img = np.array(pil_img)

        # Step 1: Preprocess image for better OCR
        img = cv2.resize(img, None, fx=3, fy=3, interpolation=cv2.INTER_LINEAR)
        gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
        blur = cv2.GaussianBlur(gray, (3, 3), 0)
        _, thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
        thresh = cv2.bitwise_not(thresh)  # Invert for dark digits

        # Step 2: Run OCR
        results = reader.readtext(thresh, detail=1)

        # Step 3: Extract numbers like 65.20 or 50
        weight_candidates = []
        for bbox, text, conf in results:
            clean = text.lower().replace("kg", "").replace("kgs", "").strip()
            if re.fullmatch(r"\d{2,4}(\.\d{1,2})?", clean):
                weight_candidates.append((clean, conf))

        if not weight_candidates:
            return "Not detected", 0.0

        # Step 4: Choose highest confidence number
        weight, confidence = sorted(weight_candidates, key=lambda x: -x[1])[0]
        return weight, round(confidence * 100, 2)

    except Exception as e:
        return f"Error: {str(e)}", 0.0