📖 OCR - Optical Character Recognition

# OCR Digital Display Reading
import cv2
import numpy as np
from PIL import Image
import pytesseract

# Enhanced OCR preprocessing
def preprocess_for_ocr(image):
    # Convert to grayscale
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    
    # Noise reduction
    denoised = cv2.bilateralFilter(gray, 9, 75, 75)
    
    # Enhance contrast
    clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8,8))
    enhanced = clahe.apply(denoised)
    
    # Binary thresholding
    _, binary = cv2.threshold(enhanced, 0, 255, 
                             cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    
    return binary

# OCR with custom config
def read_digital_display(image, display_type='led'):
    # Preprocess image
    processed = preprocess_for_ocr(image)
    
    # Configure OCR based on display type
    if display_type == 'led':
        config = '--oem 3 --psm 8 -c tessedit_char_whitelist=0123456789.-'
    elif display_type == 'lcd':
        config = '--oem 3 --psm 7'
    
    # Run OCR
    text = pytesseract.image_to_string(processed, config=config)
    confidence = pytesseract.image_to_data(processed, output_type=pytesseract.Output.DICT)
    
    # Validate and clean result
    cleaned_text = validate_reading(text, display_type)
    avg_confidence = np.mean([int(conf) for conf in confidence['conf'] if int(conf) > 0])
    
    return cleaned_text, avg_confidence

# Validation based on expected format
def validate_reading(text, display_type):
    import re
    
    if display_type == 'led':
        # Expect numeric values
        match = re.search(r'[-+]?\d*\.?\d+', text)
        return match.group(0) if match else None
    
    # Add more validation rules for other display types
    return text.strip()

# Example usage
display_roi = cv2.imread('hmi_display.jpg')
reading, confidence = read_digital_display(display_roi, 'led')
print(f"Display Reading: {reading}, Confidence: {confidence:.1f}%")