Detecting Text Regions

import cv2
import numpy as np
import pytesseract

def detect_text_regions(image_path):
    """Detect where text is located in an image."""
    image = cv2.imread(image_path)
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

    # Get bounding boxes for each detected word
    data = pytesseract.image_to_data(
        gray,
        output_type=pytesseract.Output.DICT,
        config="--psm 11"  # Sparse text detection
    )

    result = image.copy()
    text_regions = []

    for i in range(len(data["text"])):
        conf = int(data["conf"][i])
        text = data["text"][i].strip()

        if conf > 50 and text:  # Only confident detections
            x = data["left"][i]
            y = data["top"][i]
            w = data["width"][i]
            h = data["height"][i]

            text_regions.append({
                "text": text,
                "confidence": conf,
                "bbox": (x, y, w, h)
            })

            # Draw bounding box
            cv2.rectangle(result, (x, y), (x+w, y+h), (0, 255, 0), 2)

    print(f"Detected {len(text_regions)} text regions")
    for region in text_regions[:10]:
        print(f"  '{region['text']}' (confidence: {region['confidence']}%)")

    cv2.imwrite("text_regions.jpg", result)
    return text_regions

detect_text_regions("document.jpg")