thanhtunguet/sample-ocr

Sample OCR

1. Set up your environment:

   • Install Python (if not already installed).
   • Install required libraries: Pillow for image handling, pytesseract for OCR, and opencv-python for image processing.

   python3 -m venv .venv
   source .venv/bin/activate

2. Install Tesseract:

   • Download and install Tesseract from here.

3. Write the Python script:

Here’s a detailed script to get you started:

Step 1: Install the necessary libraries

pip install pillow pytesseract opencv-python

Step 2: Write the Python code

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

# Path to the tesseract executable
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'  # Update this path based on your installation

def preprocess_image(image_path):
    # Read the image using OpenCV
    image = cv2.imread(image_path)
    
    # Convert to grayscale
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    
    # Apply thresholding
    _, thresh = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
    
    # Optionally, apply additional processing like noise reduction
    # blur = cv2.GaussianBlur(thresh, (1, 1), 0)
    
    return thresh

def extract_text_from_image(image_path):
    # Preprocess the image
    processed_image = preprocess_image(image_path)
    
    # Use pytesseract to extract text
    text = pytesseract.image_to_string(processed_image)
    
    return text

if __name__ == "__main__":
    # Path to the image
    image_path = 'path_to_your_image.jpg'  # Update with your image path

    # Extract text
    text = extract_text_from_image(image_path)
    print("Extracted Text:\n", text)

Step 3: Explanation of the Code

1. Importing Libraries:

   • cv2 is OpenCV for image processing.
   • pytesseract is the Python wrapper for Tesseract.
   • PIL (Python Imaging Library) for additional image handling.
   • numpy for numerical operations (used by OpenCV).

2. Preprocessing the Image:

   • Load the image using OpenCV.
   • Convert the image to grayscale, which is often more suitable for OCR.
   • Apply thresholding to binarize the image, improving the contrast between the text and the background.

3. Extracting Text:

   • pytesseract.image_to_string() is used to perform OCR on the preprocessed image.

4. Running the Script:

   • Update the path to your image and the Tesseract executable.
   • Run the script to see the extracted text from the image.

Tips for Better Accuracy

• Experiment with different preprocessing techniques like denoising, blurring, and edge detection to improve OCR accuracy.
• Ensure the image is clear and the text is legible.
• Adjust the thresholding parameters if necessary.