OCR Mini Project

import cv2
import numpy as np
import pytesseract
from PIL import Image
import re

class DocumentReader:
    """A complete OCR document reader with pre-processing and output options."""

    def __init__(self, lang="eng"):
        self.lang = lang

    def preprocess(self, image):
        """Pre-process image for best OCR accuracy."""
        gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
        denoised = cv2.fastNlMeansDenoising(gray, h=10)
        _, binary = cv2.threshold(denoised, 0, 255,
                                  cv2.THRESH_BINARY + cv2.THRESH_OTSU)
        kernel = np.ones((1, 1), np.uint8)
        cleaned = cv2.morphologyEx(binary, cv2.MORPH_OPEN, kernel)
        padded = cv2.copyMakeBorder(cleaned, 15, 15, 15, 15,
                                    cv2.BORDER_CONSTANT, value=255)
        return padded

    def read(self, image_path, config="--psm 6"):
        """Read text from an image file."""
        image = cv2.imread(image_path)
        if image is None:
            return {"error": f"Could not load: {image_path}"}

        preprocessed = self.preprocess(image)
        raw_text = pytesseract.image_to_string(preprocessed,
                                               lang=self.lang, config=config)

        # Clean the extracted text
        cleaned_text = self._clean_text(raw_text)

        # Get word count and character count
        words = cleaned_text.split()
        chars = len(cleaned_text)

        return {
            "text": cleaned_text,
            "word_count": len(words),
            "char_count": chars,
            "lines": cleaned_text.split("\n"),
        }

    def _clean_text(self, text):
        """Remove common OCR noise."""
        # Remove multiple spaces and normalise whitespace
        text = re.sub(r" +", " ", text)
        text = re.sub(r"\n\n+", "\n\n", text)
        return text.strip()

    def save_text(self, image_path, output_path):
        """Read from image and save text to file."""
        result = self.read(image_path)
        if "error" in result:
            print(result["error"])
            return

        with open(output_path, "w", encoding="utf-8") as f:
            f.write(result["text"])

        print(f"=== OCR Complete: {image_path} ===")
        print(f"Words: {result['word_count']}")
        print(f"Characters: {result['char_count']}")
        print(f"Saved to: {output_path}")
        print("\nFirst 200 characters:")
        print(result["text"][:200])

# Use the reader
reader = DocumentReader(lang="eng")
reader.save_text("document.jpg", "extracted_text.txt")